Home

AnalyZ Solutions covers the full research workflow: from loading and cleaning data, through statistical analysis and visualisation, to qualitative thematic coding. This guide walks you through every feature in detail.

What's in the platform

Typical workflows

Quantitative workflow

Load data
→
Inspect in Data View
→
Clean & prepare
→
Describe
→
Test / model
→
Visualise & Dashboard
→
Sense Making
→
Data Canvas

Qualitative workflow

Load transcripts
→
View & read
→
Code themes
→
Explore themes
→
Analyze relationships
→
Sense Making
→
Data Canvas

Mixed-methods workflow

Load both
→
Quantitative analysis
→
Qualitative coding
→
Sense Making
→
Data Canvas
Guest mode: you can explore the platform without an account, but data is limited to 50 rows / 1 MB and sessions cannot be saved. Sign up free to remove all limits.

Creating an account

Creating an account is free and needs no payment details. You can sign up with an email address and password, or continue with a Google or GitHub account.

1
Click Create account
From the landing page or the sign-in screen, choose Create account.
2
Enter your email and a password
The password must have at least 8 characters, one uppercase letter, one lowercase letter, one number and one special character (such as ! or #). A checklist on the screen ticks off each rule as you meet it. Type the password a second time to confirm it.
3
Accept the Terms and complete the human check
Tick the box to accept the Terms of Service and Privacy Policy, and complete the short human-verification check shown on the form.
4
Verify your email
A verification link is sent to your address. Click it to activate your account. If it does not arrive, check your spam folder.
5
Sign in
Go back to the sign-in screen and enter your details. You are taken straight into the app.

Signing in

On the sign-in screen you can use email and password, or the Google or GitHub buttons.

Two-step verification

Two-step verification adds a second lock to your account. After your password, AnalyZ asks for a 6-digit code that changes every 30 seconds and appears only in an app on your phone. Even if someone learns your password, they cannot sign in without the code. It is free.

Turning it on

1
Open Account Settings
On the home screen click Settings in the top bar, then open the Authentication tab.
2
Click Set up
A QR code appears.
3
Scan the QR code
Use an authenticator app such as Google Authenticator, Microsoft Authenticator, Authy or 1Password. If you cannot scan, type the key shown under the code into the app by hand.
4
Enter the 6-digit code and click Turn on
The app shows a code for AnalyZ Solutions. Typing it confirms the setup. From your next sign-in you will be asked for a code.

Good to know

Keep at least one working authenticator. If you lose access to every registered device, you cannot pass the code step. Adding a second device now is the safest protection against this.

Forgotten password

Click Forgot password? on the sign-in screen, enter your email and click Send reset link. For privacy the screen says the same thing whether or not the address is registered. Open the link in the email, choose a new password that meets the same rules as at sign-up, and sign in again. Reset links expire after a short time, so use them soon after they arrive.

Signing out

Click Sign out at the far right of the top bar (next to your email address). You are signed out on this device and returned to the landing page. Save a session file first if you want to keep your work (see Saving Sessions).

Account Settings

Open it with the Settings button in the top bar of the home screen. It has four tabs.

TabWhat you can do there
SubscriptionSee your current plan (Free, Starter, Pro or Enterprise) and its renewal date, or the end date if you have cancelled. Usage this month shows bars for Sense Making interpretations, Dashboard interpretations, Data Canvas story generations and Data Canvas slide refinements. On the Free plan you can subscribe to Starter or Pro. On Starter you can upgrade to Pro. On a paid plan, Cancel plan stops renewal.
ProfileYour email is shown for reference and cannot be changed here. You can set your display name, organisation or institution, role (researcher, evaluator, academic, student, NGO staff, consultant or other) and country. Click Save profile.
ActivityThis month's usage as simple counts, for the same four items. Usage resets on the 1st of each month.
AuthenticationTurn two-step verification on or off and manage your authenticator devices. See the section above.

Plans and monthly allowances

Allowance per monthFreeStarterPro
Sense Making interpretations11050
Dashboard interpretations11050
Data Canvas story generations2515
Data Canvas slide refinements51550

Starter is ₹200 a month and Pro is ₹400 a month. Allowances reset on the 1st of each month. Publishing public links for dashboards and Data Canvas stories is not available on the Free plan. The analysis modules themselves are available to every registered account.

Cancelling a plan

Click Cancel plan on the Subscription tab and confirm. The plan will not renew. It stays active until the end of the period you have already paid for, then returns to Free. Partial refunds are not issued. For billing questions, email support@analyz.solutions.

Data security

WhatHow it works
Data in transitAll communication between your browser and AnalyZ is encrypted with TLS (HTTPS). Data is never sent over plain HTTP.
Data in sessionUploaded files and analysis results are held in browser memory only for the duration of your session. By default they're discarded the moment you close the tab or log out.
Local device save (opt-in)You can turn on Save on this device in the sidebar to keep a working copy in this browser's local storage across closes. This is off by default, stored only on your device (never our servers), and expires automatically after 24 hours.
Server storageAnalyZ does not store your uploaded data on any server. No copies are kept after your session ends.
Saved sessions (.analyz.json files)When you save a session, the file is downloaded directly to your own device. AnalyZ does not keep a copy server-side.
Account credentialsPasswords are never stored in plain text: only a secure hash is kept. AnalyZ staff cannot see your password.
Authentication tokensLogin sessions use short-lived JWT tokens that expire automatically. Logging out invalidates the token immediately.
For sensitive research data: Because uploaded data is processed entirely in your browser's memory and never leaves your device to a storage server, AnalyZ is suitable for use with personally identifiable information (PII) subject to your organisation's data governance policies. Always consult your institution's ethics and data protection officer if in doubt.

Guest mode vs. registered account

FeatureGuestRegistered
Data limit50 rows / 1 MBUnlimited
Qualitative files1 file / 1 MBNo limit
Session saving (.analyz.json)❌✅
Download data❌✅ (Excel, JSON, Stata, SPSS)
All modules✅✅
Account required❌✅ (free)

Everything starts in the sidebar on the left. At the top it asks What type of data do you want to analyze? with two buttons, Quantitative (spreadsheets and survey files) and Qualitative (interview and focus group transcripts). Choose the one that matches your data, then follow the matching steps below.

On a phone or narrow window the sidebar is hidden. Open it with the menu button at the top left of the screen.

Load a quantitative file

1
Click the Quantitative button
It is at the top of the sidebar. If transcripts are already loaded you can switch back and forth without losing either.
2
Click Upload, or drag your file onto it
Accepted: CSV, Excel (.xlsx and .xls), Stata (.dta), SPSS (.sav), JSON, and a saved AnalyZ session (.analyz.json).
3
Wait for the green strip
Once the file is read, the sidebar shows the file name with its rows and columns count. Large files, and Stata or SPSS files, can take several seconds.
4
Pick where to go next
Click a card on the main screen, or open Modules in the top bar. If you are unsure which analysis fits, open Guided Analysis in the top bar.
5
Check the data in Data View
Open Data View from the top bar and look at the first rows before analysing anything. See Data View.
Loading a second file replaces the first. If a dataset is already loaded, AnalyZ asks Replace existing data? before it does so. Click Replace & continue to go ahead. Save your work first if you want to keep it (see Saving and Restoring Sessions).

What each file type brings with it

File typeVariable labels and value labelsNotes
CSVNot includedAdd labels afterwards in Data View, using Edit variable labels, or by right-clicking a column header and choosing Variable & Value Labels.
Excel (.xlsx, .xls)Optional, from a second sheetAn Import Excel data window opens first (see below), where you choose the sheet, the row with variable names and, optionally, a sheet of labels.
Stata (.dta)ImportedA Select variables to load window opens first (see below).
SPSS (.sav)ImportedSame variable picker as Stata. The file is read by AnalyZ's server, so see the privacy note below.
JSONNot includedShould be a list of records, one per row.
.analyz.jsonIncludedRestores only the dataset half of the file here. To restore everything use Restore Session instead.

Excel files: the Import Excel data window

Every Excel file opens a window titled Import Excel data, even when it has only one sheet. It shows the file name and, for workbooks with several sheets, how many there are.

1
Choose the Sheet
Only shown when the workbook has more than one sheet. Click the round button beside the sheet you want. Each sheet shows its rows and columns, and empty sheets are greyed out.
2
Tell AnalyZ where the variable names are
The box Variable names are in row is filled in with a guess. A preview of the first rows shows the chosen row highlighted. Rows above it are skipped, and data starts on the next row. If the sheet has no names, tick No variable names (use Column_1, Column_2, …).
3
Add labels from another sheet (optional)
If the workbook has several sheets, Variable labels (optional) lets you set Labels are in sheet. Then choose Variable names in column and Labels in column, and tick First row is a header if the labels sheet has a title row. A line such as age → Age of respondent shows what will be matched. Names are matched to your data columns ignoring capital letters.
4
Click Load data
The sidebar then reports how many variable labels were applied. If none matched, it says so. Cancel or ✕ closes the window without loading.
A sheet cannot be both the data and the labels. Choosing a labels sheet that is the same as the data sheet clears the labels choice.

Stata and SPSS files: choosing variables

Survey files often hold hundreds of variables. After you choose a .dta or .sav file, a window titled Select variables to load opens.

1
Search or tick variables
Type in Search by variable name or label…, then tick the variables you need. Select all and Clear all do what they say, and Select first 50 is a quick start.
2
Mind the limit
If you choose too many, the message Deselect some before loading appears and the Load button stays off. Keep only what you will analyse.
3
Choose a label language
If the file stores labels in more than one language, the Label language: menu lets you pick one.
4
Load
Click the Load N variables button at the bottom of the window, where N is the number you ticked.
Privacy. Most files are read inside your browser and are not uploaded. These steps do send data to AnalyZ's server so it can be processed: reading an SPSS (.sav) file, extracting text from PDF and Word (.docx) transcripts when you upload them, exporting Stata (.dta) or SPSS (.sav), and fetching data through Connect API. Nothing is kept after the request finishes. Do not use these routes for data your organisation does not allow you to process outside your device.

Connect to a survey platform (Connect API)

Use this to pull data straight from a survey tool instead of exporting a file. It needs an account (it is greyed out for guests with the text Connect API, sign up to unlock).

1
In the sidebar, click Connect API
It sits below Upload. A list headed Choose a platform: opens.
2
Choose a platform
KoBo, ODK Central, REDCap, Google Sheets (public CSV link) or a custom API. ← Back returns to the list.
3
Fill in the form
Give it a Connection name (for example Survey 2024) and the API URL. For a custom API you can also set how the key is Sent as, for example an Authorization: Token header, and an optional JSON data path (leave blank to auto-detect).
4
Load
Click Connect & fetch data. The text changes to Connecting… while it works. If a dataset is already loaded you will be asked Replace existing data?
5
Refresh later
Open Connected sources in the sidebar. Click the refresh button on a source to pull the newest records, or ✕ to remove it.

KoBo Toolbox in detail

1
Click Connect API, then choose KoBo
The panel is titled Connect to KoBo.
2
Paste any KoBo link
Under Paste the form URL from your browser, paste the link of your form or just the server address, for example https://kf.kobotoolbox.org. If no form is recognised, your project list appears so you can choose one.
3
Choose how to sign in
How would you like to sign in? gives two options. API Token: click Open token page, copy the token shown, paste it under Paste your token. Username & Password: enter your KoBo username and password. They are used once to get a token and are never stored.
4
Choose variables
A Select variables to load window lists every question with its label. Pick what you need and click Load N variables.
5
Sign-in expired?
When you refresh a source after your session expired, the prompt Your session has expired. Reconnect to refresh. appears. Paste the token again or choose Username & Password.
Tick Load as qualitative transcript on the connection form if the source holds interview text rather than a data table. The source then appears with (qual) beside its name.
A saved session file keeps the connection name, URL and platform. It never keeps your token or password.

Try a demo dataset

Under Or try a demo dataset: in the sidebar there are two buttons, Test data without labels and Test data with labels. Both load a 300-row example survey with age, income, gender, education, score, region and employment. The second one already has labels, so it is the better choice for following this guide.

Load interview transcripts

1
Click the Qualitative button
At the top of the sidebar. If a quantitative dataset is loaded it stays in memory.
2
Click Upload, or drag your files onto it
Accepted: plain text (.txt), Word (.docx), PDF (.pdf) and CSV (.csv). Text is extracted from Word and PDF files automatically. A CSV that looks like a table of numbers is treated as a dataset instead of a transcript.
3
Check the count
The sidebar lists each transcript you loaded.
4
Open the module
On the home screen choose the Qualitative Analysis cards, or use Modules in the top bar. See Qualitative Analysis.
Markdown (.md) files are not offered by the file picker. Rename them to .txt. Guests can load one file of up to 1 MB. Signed-in users have no file limit.

No transcripts yet? Under Or try a demo transcript: click Try dummy transcript. It loads a sample interview so you can practise coding.

To save transcripts, click Download Transcripts in the sidebar, then choose Word document (.docx), Plain text (.txt) or Markdown (.md). Several transcripts download together as a .zip.

Clear, undo and redo

Button (sidebar)What it does
Clear DatasetRemoves the loaded dataset and its results from the app.
Clear TranscriptsRemoves all loaded transcripts.
✕ Clear allRemoves the demo data or everything loaded, depending on where it shows.
Undo and RedoStep backwards or forwards through changes you made to the dataset in Data Management or Data View, such as recodes, deletions or sorting.
Clearing cannot be undone with Undo. Save a session first if you may need the work again.

Top navigation bar

The bar along the top of every page holds the main controls. Some buttons appear only where they are useful.

ButtonWhat it doesWhere it appears
HomeReturns to the first home screen with all the choices. If you are already on that screen, it takes you to the AnalyZ website.Everywhere
Data ViewerOpens a panel on the right showing your dataset, so you can look at the data and make quick fixes without leaving the page you are on. See below.Data Management, Descriptive, Inferential, Scale & Reliability and Data Visualization
ModulesA drop-down list of every module, grouped as Quantitative Data, Qualitative Data and Tools. Point at a module that has sections (for example Inferential Analysis) and a second panel lists them, so you can jump straight to, say, Regression or Mind Map.Inside any module, and on the home screen once you have chosen Quantitative or Qualitative
Guided AnalysisOpens the assistant that recommends an analysis from your goal. See below.Everywhere
HelpOpens this guide in a new tab.Everywhere
SettingsOpens Account Settings (subscription, profile, activity, authentication).Home screen only
Your email and Sign outShows who is signed in. Sign out ends your session on this device.Everywhere

Data Viewer

The Data Viewer is a read-and-fix window onto your dataset. It does not replace Data View, but it is handy while you are partway through an analysis.

Right-click a column heading to open its quick menu: Variable Type, Variable & Value Labels, Define Missing Values, Rename, Recode, Delete, Duplicate Column, Split Column, Encode Variable, Sort Ascending or Descending, Hide Column and View Frequency. Each opens a small panel for that one task. The full versions are in Data Management.

Home screen

The first home screen asks What do you want to explore today? and offers six cards:

CardOpens
SamplingSample size estimation, power analysis and drawing a sample. No dataset is needed.
Quantitative AnalysisThe quantitative module grid, with the sidebar for loading data.
Qualitative AnalysisThe qualitative module grid, with the sidebar for loading transcripts.
Data VisualizationThe chart builder and dashboards.
Sense MakingAI-assisted synthesis across your saved analyses.
Data CanvasThe story builder for shareable findings.

Once you are in a quantitative or qualitative mode, the left sidebar is where you load data, save or restore sessions and download data. The module cards are on the main screen. Cards that need data are greyed out and say Upload your data first (or Upload transcripts first) until something is loaded. Compare Coding is the exception, because it works from files you load inside it.

Quantitative modules

ModuleWhat it does
Data ViewInspect your dataset, variable types, labels and missing values.
Data ManagementCreate, recode, rename, delete and reshape variables. Merge and append datasets.
Descriptive AnalysisFrequencies, summary statistics, cross-tabulations, tables and data quality.
Inferential AnalysisCompare means and percentages, correlation, regression (including multilevel) and non-parametric tests.
Effect AnalysisEstimate a programme's effect with propensity score matching, difference-in-differences, regression discontinuity or instrumental variables.
Data Visualization28 chart types and multi-chart dashboards with cross-filtering.
Scale & ReliabilityCronbach's alpha, PCA and exploratory factor analysis.
Structural Equation Modelling (Beta)Confirmatory factor analysis, structural models, group comparison, path analysis and mediation.

Qualitative modules

ModuleWhat it does
View TranscriptsRead, edit and tag transcripts. Overview stats, word cloud and word search.
Thematic CodingBuild a codebook, assign codes to passages and review coded segments.
Compare CodingLoad two coders' exports and compute Cohen's kappa.
Explore ThemesSaturation, theme comparison, theme connectedness and segment length.
Analyze RelationshipsNetwork Map, Mind Map, Thematic Clusters, Code Sequence, Configurational Analysis, QCA and Persona Builder.

Tools

ModuleWhat it does
SamplingSample size estimation, power analysis and drawing a sample from a sampling frame.
Sense MakingAI synthesis of the analyses you have saved, quantitative and qualitative together, into one narrative.
Data CanvasA scrollable story of charts, tables and quotes, shareable as a link.

Guided Analysis wizard

Available at any time from the Guided Analysis button in the top navigation bar. It helps you identify the right analysis when you are unsure where to start, or when you want confirmation that you are using the right approach for your goal.

The wizard works in three stages:

1
Describe your goal
Type your goal in plain language, for example, "I want to find out if men and women scored differently on the satisfaction question." The AI interprets your goal, identifies the relevant analysis, and pre-selects your variables from the loaded dataset. Or skip the text box and pick from the goal list directly.
2
Confirm the interpretation and select variables
If you used the free-text option, the wizard shows how it interpreted your goal and asks you to confirm. It then shows relevant variable dropdowns populated from your actual dataset, group variable, outcome variable, etc., pre-filled where it could infer them from your description.
3
Review the recommendation and go
The wizard shows the recommended analysis, the specific module and test, and a plain-language explanation of why it fits your goal. Click Take me there to open the module directly.

Goals covered include: describing data, comparing groups, predicting an outcome, exploring relationships, validating a scale, building a chart or dashboard, cleaning data, calculating sample size, understanding qualitative data, finding patterns across transcripts, mixed-methods analysis, and producing a data story.

The AI never sees your raw data, it only uses your column names and types to make variable suggestions. You always review and confirm before proceeding.

Sub-sample filtering

Most quantitative analysis modules have a 🔍 Filter data panel at the top. Expand it to add one or more conditions (e.g. gender == Female, age > 30). All analysis in that module will run on the filtered subset only. An info banner confirms the active filter. Remove conditions to return to the full dataset.

Guests cannot save sessions. You will see Sign up free to save sessions. in the sidebar. Create a free account to use these buttons.

Save a session file

1
Load your data or transcripts
Quantitative data, qualitative transcripts, or both.
2
Click Save Session in the sidebar
A file ending .analyz.json downloads to your computer.
3
Store the file safely
Keep it with your project files. Treat it like the data itself: it contains your dataset and transcripts.

The file holds: your dataset, variable and value labels, variable types, survey design, transcripts, codebook, coded segments, notes, connected API sources (without credentials) and saved dashboard charts.

Restore a session

1
Click Restore Session in the sidebar
It works from either Quantitative or Qualitative mode.
2
Choose your .analyz.json file
Everything returns as you left it, in both modes.
3
Refresh API sources if needed
Open Connected sources and click the refresh button, then paste your token again.
If you drop a .analyz.json file onto the ordinary Upload button instead, only the half that matches the current mode is restored. Use Restore Session to get everything.

Keep your work on this device (optional)

By default everything is lost when you close the tab. You can ask AnalyZ to keep a working copy in your browser.

1
Answer the prompt
The first time data is loaded, Save this session on this device? appears. Click Save it or Don't save.
2
Change your mind later
Use the Save on this device switch in the sidebar. Turning it off clears what was saved.
3
Know the limit
A saved copy expires after 24 hours. For anything longer, use Save Session and keep the file.

This copy is stored only in your browser and is not sent to AnalyZ's servers.

Download your data

Download Data in the sidebar saves the current dataset, including any recodes and new variables. Choose a format: Excel (.xlsx), JSON (.json), Stata (.dta) or SPSS (.sav). There is no CSV option here. Stata and SPSS files include variable and value labels. These two are prepared by AnalyZ's server, which handles the data only for the duration of the request.

The right method depends on two things: what you are asking, and what kind of variables you have (see Statistics Basics if the types are unfamiliar). Use the tables below, then follow the link to the full explanation. You can also open Guided Analysis from the top bar and answer a few questions to be pointed to a method.

I want to describe my data

My questionUseWhere
Is my data clean enough to analyse?Data QualityDescriptive Analysis
How many people gave each answer?FrequencyDescriptive Analysis
How do two categorical questions line up?Cross-TabulateDescriptive Analysis
I need a standard 'Table 1' for a reportTableDescriptive Analysis
What is the typical value and how spread out is it?Summary StatisticsDescriptive Analysis

I want to compare groups or time points

Work from left to right: how many groups, whether they are the same people measured repeatedly, and whether the numbers are roughly bell-shaped. If you are unsure about the shape, the rank-based test is the safer option.

Groups being comparedNumeric outcome, roughly bell-shapedOrdinal, skewed or small sample
One group against a fixed valueOne-sample t-testNo direct equivalent here. Report the median and use Frequency.
Two separate groupsIndependent t-testMann-Whitney U
Same people measured twicePaired t-testWilcoxon Signed-Rank
Three or more separate groupsOne-way ANOVAKruskal-Wallis
Two factors at once (for example sex and region)Two-way ANOVANo direct equivalent. Use regression with both variables.
Same people measured three or more timesRepeated-Measures ANOVAFriedman Test

For percentages instead of averages: use the Chi-square test to compare a category breakdown between groups, the Z-test for one proportion to compare a percentage with a target, and the Z-test for two proportions to compare two percentages.

I want to know whether two things are related

The two variables areUse
Both categoricalChi-square test (or Cross-Tabulate for a quick look)
Both numeric, roughly straight-line, no extreme outliersPearson Correlation
Ordinal, skewed, or with outliersSpearman Correlation
One numeric, one categoricalCompare groups (see above)

I want to predict or explain an outcome

Regression lets you look at several predictors at once, so each one's effect is shown while the others are held steady. The method is chosen by the type of outcome.

My outcome isExampleUse
A numberIncome, test score, weightLinear Regression
Yes or noAttended, employed, recoveredLogistic Regression (or Probit)
Three or more unordered categoriesWhich programme was chosenMultinomial Regression
Ordered categoriesSatisfaction from 1 to 5, severityOrdinal Regression
A count of eventsNumber of visits or incidentsPoisson or Negative Binomial Regression
Any of the above, but people are grouped in schools, clinics or villagesScores of pupils in schoolsMultilevel Regression
A number, and you want to see how much each set of predictors addsDemographics first, then programme variablesHierarchical Regression

I want to know whether a programme caused a change

Your situationUse
One round of data, no random assignment, rich background information on everyonePropensity Score Matching
Programme and comparison groups measured before and afterDifference-in-Differences
Eligibility decided by a cut-off score, age or incomeRegression Discontinuity Design
Take-up is tied up with unmeasured factors and you have a credible instrumentInstrumental Variables
You want to know how an effect happens, through which middle stepMediation or Counterfactual Mediation

I want to check a questionnaire scale

My questionUse
Do these items hang together as one scale?Cronbach's Alpha
How many underlying factors do my items form? (I have no theory yet)Exploratory Factor Analysis
I just want fewer variables or a combined scorePrincipal Component Analysis
Does my planned structure of items and factors fit the data?Confirmatory Factor Analysis
Does the scale work the same way in every group?Compare Groups (measurement invariance)

I am planning a study

My questionUse
How many people do I need to survey?Sample Size Estimation
Will my planned sample be big enough to detect a difference?Power Analysis
I have a list of units and need to pick the sampleDraw a Sample

You do not need a statistics background to use AnalyZ. Every result comes with a written interpretation. This page explains the ideas those interpretations rely on, so you can check them against your own judgement and explain them to others. Each method in the guide links back here when it uses one of these ideas.

Sample and population

The population is everyone you want to learn about, for example all households in a district. The sample is the smaller group you actually measured, for example 400 of those households. Almost every test in AnalyZ asks the same underlying question: could what I see in my sample just be luck of who happened to be picked, or does it probably hold for the whole population?

This is why a bigger sample makes results more trustworthy, and why a result from 12 people deserves more caution than the same result from 1,200.

Types of variables

Choosing the right method starts with knowing what kind of variable you have. AnalyZ guesses the type for you (a number column with more than 10 different values is treated as continuous, otherwise as categorical) and you can change it in Data Management, under Variable Types.

TypeWhat it isExamplesUsually analysed with
Continuous (numeric)A measured number where differences are meaningfulAge, income, test score, weight, time in minutesMeans, correlation, linear regression, t-tests, ANOVA
BinaryTwo possible answersYes/No, male/female, passed/failed, treated/not treatedPercentages, chi-square, logistic regression
NominalThree or more categories with no natural orderRegion, religion, type of programme, occupationFrequencies, cross-tabulation, chi-square, multinomial regression
OrdinalCategories with a natural order but unequal gapsSatisfaction from 1 to 5, education level, severity (mild, moderate, severe)Medians, rank-based tests, Spearman, ordinal regression
CountWhole numbers of events, never negativeNumber of clinic visits, number of incidents, children in a householdPoisson or negative binomial regression
A 5-point satisfaction question is ordinal, but many researchers treat it as numeric when it is part of a longer scale that is added up into a total score. For a single Likert question on its own, the rank-based tests are the safer choice.

Significance and p-values

Most tests begin by assuming that nothing is going on: the two groups are really the same, the variables are really unrelated, the programme really did nothing. This starting assumption is called the null hypothesis. The test then asks how surprising your data would be if that assumption were true.

The p-value is the answer to that question, as a number between 0 and 1. A small p-value means your data would be very unlikely if nothing were going on, so you take the result seriously. A large p-value means the pattern you see is the kind of thing chance produces easily.

Marker in the tablesMeaningHow AnalyZ words it
***p-value below .001significant at the 0.1% significance level
**p-value below .01significant at the 1% significance level
*p-value below .05significant at the 5% significance level
nsp-value .05 or aboveno significant difference or relationship

The significance level (5% is the usual choice) is the line you draw in advance. Choosing 5% means you accept being fooled by chance about 1 time in 20.

A p-value is not the probability that your finding is true, and it says nothing about how big or important an effect is. With a large sample, tiny and unimportant differences become significant. With a small sample, large and important differences can fail to reach significance. Always read the effect size and the confidence interval next to it.

Confidence intervals

A single number from a sample, such as an average of 72, is only an estimate. A 95% confidence interval gives a range, for example 69 to 75, that is built so that it would capture the true population value in 95 out of 100 repeats of the study. Wide intervals mean you know little. Narrow intervals mean the estimate is precise.

Intervals also give you a quick significance check. If the interval for a difference does not include 0, or the interval for a ratio (odds ratio, rate ratio) does not include 1, the result is significant at the 5% level.

Effect sizes

An effect size answers how big, while a p-value answers whether it is distinguishable from chance. AnalyZ reports an effect size alongside most tests. These are the rules of thumb it uses to describe them in words:

Effect sizeUsed withSmallMediumLarge
Cohen's dt-tests (difference in means)0.20.50.8
Eta-squared and partial eta-squaredANOVA, Kruskal-Wallis0.010.060.14
Correlation r or rhoCorrelation0.10.30.5
Effect size rMann-Whitney U, Wilcoxon0.10.30.5
Cramér's VChi-square (2 by 2 table; larger tables need smaller cut-offs)0.10.30.5
R-squaredRegression (share of variation explained)0.020.130.26
Odds ratioLogistic regression1.5 (or 0.67)2.5 (or 0.40)4.3 (or 0.23)

These are conventions, not laws. In some fields a correlation of 0.2 is valuable, and in others 0.7 would be disappointing. Judge size against what is normal in your topic.

Assumptions in plain terms

Every test works properly only when your data behave in certain ways. You do not need to memorise the mathematics, only the four ideas below.

AssumptionIn plain wordsCheck it byIf it fails
Roughly bell-shaped data (normality)Most values sit near the middle, with fewer and fewer further away. Matters most for small samples (under about 30 per group).A histogram in Data Visualization, and the skewness value in Summary StatisticsUse the rank-based alternative under Non-parametric Tests
Similar spread in each group (equal variance)The groups are equally scattered around their averages.Compare the SDs in the group table, and the Levene p-value on the independent t-testInterpret with caution, or use Mann-Whitney U
Independent observationsOne person's answer does not influence another's. Broken by repeated measures on the same person or by people grouped in the same village or school.Think about how the data were collectedUse a paired or repeated-measures test, or Multilevel Regression
Straight-line relationship (linearity)Each extra unit of X changes Y by about the same amount.A scatter plot in Data VisualizationTransform the variable, or use Spearman for monotone patterns
Large samples (hundreds of cases) are forgiving of mild departures from a bell shape. Outliers are a different matter, because a few extreme values can drag averages around. Run Data Quality first to find them.

Correlation and causation

A relationship between two variables does not show that one causes the other. Children with bigger shoe sizes read better, but both depend on age. Regression can adjust for other variables you measured, which helps, but only designs that build a fair comparison group can support a claim about cause. That is the purpose of the methods in Effect Analysis (matching, difference-in-differences, regression discontinuity and instrumental variables) and of Counterfactual Mediation in Structural Equation Modelling.

Reading a results page

Every analysis page in AnalyZ follows the same layout, so once you know one you know them all.

PartWhat it gives you
Analysis set upWhere you choose variables and options. The Run or Generate button sits at the bottom, next to the Apply survey weight tick box when a survey weight has been set.
TablesThe numbers: group statistics, the test statistic, the p-value with its stars, and effect sizes. Each table can be exported.
ChartsWhere a picture helps (means with error bars, forest plots, scree plots). Charts can be edited and downloaded.
InterpretationA short written reading of the main result in plain language, suitable as a starting point for your report. It names the groups and variables using your labels.
Export and share buttonsCSV, Excel and Word downloads, Add to Canvas (a slide for your data story) and Add to Sense Making (sends the result for AI synthesis).

Type in the box to narrow the list. Terms are written the way they appear in AnalyZ tables and interpretations.

TermIn plain wordsWhere you meet it
Alpha (Cronbach's)A number from 0 to 1 showing how well a set of questionnaire items hang together as one scale.Scale & Reliability
Alpha (significance level)The cut-off you set for calling a result significant. 0.05 means 5%.All tests
ANOVAA test that compares averages across three or more groups at once.Inferential
AssumptionA condition your data should roughly meet for a test to be trustworthy.All tests
ATT, ATC, ATEThe programme effect on those treated, on those not treated, or on everyone.Effect Analysis
Average variance extracted (AVE)How much of an item set's variation is captured by its factor rather than by noise. 0.50 or more is good.SEM
Balance (covariate balance)Whether two groups look alike on background characteristics.Propensity Score Matching
BandwidthHow close to a cut-off a case must be to be used in the comparison.Regression Discontinuity
Beta (B, coefficient)In regression, how much the outcome changes for a one-unit increase in a predictor.Regression
Bonferroni correctionA way of making p-values stricter when many comparisons are made at once.Non-parametric, RM ANOVA
BootstrapRe-running an analysis on many random re-draws of your data to see how much the answer wobbles.Effect Analysis, SEM
Categorical variableA variable whose values are groups or labels.All
CFI and TLIFit measures for a factor or structural model. 0.95 or more is good.SEM
Chi-square (χ²)A measure of how far observed counts are from what would be expected if nothing were going on.Chi-square, SEM fit
Cohen's dThe size of a difference between two means, in standard deviations.t-tests
Confidence intervalA range that likely contains the true population value.All
ConfounderA third factor that influences both the thing you are studying and the outcome, and can fake or hide an effect.Effect Analysis
Correlation (r)A number from -1 to +1 showing how closely two variables move together.Correlation
CovariateAn extra variable you adjust for in an analysis.Regression, Effect Analysis
Cramér's VThe strength of a link between two categorical variables, from 0 to 1.Chi-square, Cross-Tabulate
Degrees of freedom (df)A technical count that depends on sample size and number of groups. It is needed to look up the p-value, and you rarely need to interpret it.All tests
Dependent variableThe outcome you are trying to explain or predict.Regression, ANOVA
Design effect (DEFF)How much larger a sample must be because people were chosen in clusters rather than at random.Sampling
Dummy variableA 0/1 variable standing for one category. AnalyZ creates these for you.Regression
E-valueHow strong an unmeasured confounder would need to be to explain a result away.Effect Analysis, SEM
Effect sizeHow big a difference or relationship is, whether or not it is significant.All
EigenvalueHow much of the total variation in your items a component or factor captures. Above 1 is the usual retention rule.PCA, EFA
EndogenousA variable that is itself influenced by other things in the model, or by unmeasured factors.Instrumental Variables
F statisticThe ratio of differences between groups to differences within groups. Larger means stronger evidence of a group difference.ANOVA, regression
Factor (statistical)A hidden trait, such as 'anxiety', that is assumed to drive answers to several questions.EFA, CFA
Factor loadingHow strongly an item reflects its factor, on a scale where higher is stronger.EFA, CFA, PCA
HeapingRespondents rounding to favourite numbers, such as ages ending in 0 or 5.Data Quality
Hypothesis (null)The starting assumption that nothing is going on.All tests
ICC (intraclass correlation)The share of variation that sits between groups rather than within them.Multilevel, Sampling
Independent variable (predictor)A variable you use to explain or predict the outcome.Regression
InstrumentA variable that affects take-up of a treatment but affects the outcome only through that treatment.Instrumental Variables
InteractionWhen the effect of one variable depends on the level of another.Two-way ANOVA, regression
InterceptThe predicted outcome when all predictors are zero.Regression
KMOA check that your items share enough common variation for factor analysis. Above 0.60 is acceptable.PCA, EFA
Latent variableA trait that cannot be measured directly and is inferred from several questions.SEM
Likert scaleA set of ordered answer options such as strongly disagree to strongly agree.Surveys
Margin of errorThe plus-or-minus range around a survey estimate.Sampling
Mean, median, modeThe average; the middle value when sorted; the most common value.Descriptive
MediatorA middle step through which one variable affects another.SEM
MulticollinearityPredictors that are so closely related that their separate effects cannot be told apart.Regression
Non-parametric testA test that works on ranks and so does not require bell-shaped data.Non-parametric Tests
Odds ratio (OR)How many times higher (or lower) the odds of an outcome are for each one-unit increase in a predictor. 1 means no effect.Logistic regression
OutlierA value far from the rest, which may be an error or a genuine extreme case.Data Quality
p-valueHow surprising your data would be if nothing were going on. Small means surprising.All tests
Parallel trendsThe assumption that without the programme both groups would have changed by the same amount.Difference-in-Differences
Power (statistical)The chance that a study detects a real effect of a given size. 80% is the usual target.Sampling
Propensity scoreA person's estimated chance of being in the programme given their characteristics.Propensity Score Matching
R-squared (R²)The share of the outcome's variation that your predictors explain.Regression
Random effectA group-specific difference, such as a school's own baseline, that is allowed to vary.Multilevel Regression
Reference categoryThe group everything else is compared against.Regression
ReliabilityHow consistently a measure gives the same answer.Scale & Reliability
RMSEA and SRMRMisfit measures for a model. Lower is better. RMSEA 0.06 or less and SRMR 0.08 or less are good.SEM
Running variableThe score or measure whose cut-off decides who gets a programme.Regression Discontinuity
Sampling frameThe list from which a sample is drawn.Sampling
Scree plotA line chart of eigenvalues used to decide how many factors to keep.PCA, EFA
SkewnessWhether a distribution leans to one side. Positive skew has a long tail of high values.Summary Statistics
Standard deviation (SD)The typical distance of values from the average.Descriptive
Standard error (SE)How much an estimate would vary from sample to sample.All tests
Standardised effectAn effect expressed in standard deviations so that different variables can be compared.SEM, regression
Straight-liningGiving the same answer to every item in a grid.Data Quality
Treatment and comparison groupThe people who received a programme and those who did not.Effect Analysis
VarianceThe SD squared. Used inside many tests.Descriptive
Weak instrumentAn instrument that barely predicts take-up, making results unreliable.Instrumental Variables
z-scoreHow many SDs a value lies from the mean.Tests of proportions

Data View is your first stop after loading a file. Use it to confirm everything loaded correctly and to get familiar with your variables before running any analysis.

The data table

Use the row range slider at the top of the controls bar to set which rows to display. Drag the left handle to set the start row and the right handle to set the end row, for example, drag to rows 100–300 to inspect a specific section of a large dataset. The label beside the slider shows the active range (e.g. 1–50). A missing value indicator appears automatically, a green badge if the dataset is complete, or an orange warning with a count if there are missing cells.

Showing value labels

Click Show value labels above the table. Any cell with a label mapping will display as 1 (Male) instead of just 1. Click again to return to raw values. When a dataset has value labels (e.g. imported from Stata or SPSS), this toggle is switched on automatically: so you see decoded labels by default and can turn them off to inspect raw codes.

Column details

Expand the Column details section below the table for a full inventory of every variable:

FieldWhat it shows
LabelHuman-readable variable description (if one has been added)
Value labelsMappings e.g. 1 = Male, 2 = Female
TypeData type (int64, float64, object, etc.)
Non-nullCount of rows with a value
NullsCount of missing values
UniqueNumber of distinct values
SampleThe first non-missing value in the column

Editing variable labels

Expand Edit variable labels at the bottom of the page. Select a variable, type a descriptive label (e.g. "Age of respondent in years"), and click Save label. Labels appear throughout the app wherever that variable is used. Leave the field blank and click Save to clear a label.

Good variable labels make output tables and charts far easier to read, especially when sharing results with others. Add them before running analysis.

Column menu

Right-click a column header to open a menu of quick actions for that column, without leaving Data View.

Menu itemWhat it does
Variable TypeSwitches the column between continuous and categorical. The change applies in every analysis module.
Variable & Value LabelsAdds or edits the column's label and the labels for its coded values.
Define Missing ValuesMarks coded values, such as 99 for refused or 999 for don't know, as missing, so analyses treat them like blanks.
Survey DesignOpens the survey design form. See the Survey weight section below.
Rename Variable, Recode Variable, Delete VariableThe same operations as in Data Management, applied to this column.
Duplicate Column, Split Column, Encode VariableThe same operations as in Data Management, applied to this column.
Sort Ascending, Sort DescendingSorts the rows by this column. The row order changes in the dataset, and the Undo button in the controls bar restores it.
View FrequencyShows a frequency table of the values in this column.

Column Inspector

Click any column header in the data table to open the Column Inspector: a slide-in panel on the right side of the screen. It shows the full distribution of that column alongside key statistics, so you can explore your data without running a formal analysis.

Click a different column header to switch instantly. Close the panel with the ✕ at its top. The Inspect tab on the right edge of the screen is always visible and lets you open or hide the drawer at any time.

Categorical columns

The inspector shows a horizontal bar chart with one bar per distinct value, sorted by frequency. Each bar displays the count and percentage inline. Click any bar to filter the data grid to rows containing that value, or exclude it. Reset removes the filter; Exclude all hides every value at once. Filters applied here are view-only and do not modify your dataset.

Numeric columns

The inspector shows a binned histogram. Use the Bins slider to adjust the number of bins (5 to 40) and reshape the distribution in real time. A yellow dashed line marks the mean and a blue dashed line marks the median, when they diverge noticeably, the distribution is skewed. Key statistics are shown above the chart: valid count, missing count, unique values, range, mean, median, and standard deviation.

Changing variable type from the inspector

The inspector includes a Variable type row showing the current type, continuous or categorical: with an (auto) suffix when the type has been inferred rather than set explicitly. Click Change to expand the type editor inline. Once saved, the type badge updates immediately and the distribution chart adjusts to reflect the new classification. The change propagates to all analytical modules, Descriptive Analysis, Inferential, Visualization, for the remainder of the session.

Variable type overrides are saved as part of your session. If you have Save on this device turned on, they survive a tab refresh or navigate-away. They are also included when you download a .analyz.json session file.

Column menu

Right-click any column header in the data table for quick actions on that variable, without leaving Data View:

Moving, hiding and reordering columns

These change only what you see in Data View. Your data and analyses are not affected.

Filtering the table

Each column header has a small funnel icon. Click it to open a value filter: type in Search values…, tick or untick the values you want to see, and use Select all or Clear. Remove this filter clears that column, and Clear filters (n) in the controls bar clears all of them. Blank cells appear as (Blank). Filters only change what the table shows. They never alter or delete data, and the missing-value count always reflects the full dataset.

Survey weight

If your data come from a survey with sampling weights, declare them once and the analysis modules will use them. Click Survey design in the Data View controls bar, or right-click a column header and choose Survey Design. The same form opens from the Set survey weight button beside Generate on the analysis pages.

FieldWhat to enter
Weight variableThe numeric column that holds each respondent's sampling weight. This is the only field you need.
Divide weight by (optional)A number that every weight is divided by, without changing your data. DHS weights are stored multiplied by 1,000,000, so enter 1000000 for them. A link in the form fills it in.
Strata (optional)The variable that identifies the strata of the sample design.
Cluster or PSU (optional)The variable that identifies the primary sampling units, for example villages or enumeration areas.

Before you apply, the form shows how many rows can be used, the weighted sample size, the design effect from the weights, the weight range, and the number of strata and clusters. Rows with a missing or zero weight are left out of weighted results. Once a design is set, the controls bar button shows the weight variable, and the variables in the design carry a small W, S or C badge in their column headers.

The survey design is saved with your session file (.analyz.json) and with Save on this device. Loading a new file starts without a design, and deleting a variable that the design uses removes the design.

On the analysis pages, an Apply survey weight option beside Generate is ticked by default. Untick it to see the unweighted result. How each analysis changes is described under Descriptive Analysis and Inferential Analysis. Data Quality does not use survey weights.

Data Management gives you a full set of data preparation tools without writing any code. A live row / column / missing-value count is shown at the top of each action screen. Every action is undoable (up to 20 steps).

How to use any Data Management tool

1
Open the module
Click Modules in the top bar, then Data Management under the quantitative tools. Or click the Data Management card on the home screen. You need a dataset loaded first.
2
Click a tool card
Each card is one tool, listed in the table below. The tool opens on its own screen with a ← Back link to return to the cards.
3
Fill in the form
Choose variables from the drop-down menus, most of which have a Search variables… box. A preview shows the first rows of the result before you apply it.
4
Click the action button
The button names are in the table below. A green message confirms what changed, and the row and column counts at the top update.
5
Undo if needed
Click ↩ Undo at the top of the screen, or Undo in the sidebar. Up to 20 steps can be undone.
Tool cardButton to clickTypical use
Variable & Value LabelsDownload label template (.csv), then Upload filled template (.csv); or the manual editor and Apply labelsGive variables readable names and show 1 as Male.
Encode Text VariableEncode selected columnTurn Yes/No or Male/Female text into numeric codes.
Variable TypesPick an Override for a variable; Reset all overrides undoes themTreat a 1 to 5 rating as categorical or continuous.
Sort DataChoose Sort by, add Then by rules, applyOrder rows by region, then by age.
Duplicate ColumnDuplicate ColumnKeep a safe copy before recoding.
Create VariableCreate VariableTotal score, ratios, pass/fail flags.
Recode VariableApply RecodeGroup ages, reverse-score items, collapse categories.
Rename VariablesRename, Add Prefix or Add SuffixTidy names. Spaces are not allowed.
Delete VariablesDelete Selected VariablesDrop columns you will not use.
Delete ObservationsDelete RowsRemove rows by number (for example 3, 10-15) or by a condition such as age < 18.
Manage DuplicatesRemove DuplicatesDrop repeated records, keeping the first or the last.
Merge DatasetsUpload second dataset, add Key variables, choose Join type, then Merge DatasetsAdd household data to individual data.
Append DatasetUpload a second dataset, then Append DatasetsStack survey waves.
Split ColumnPreview split, then Split ColumnBreak 'First Last' into two columns, or split multi-select answers into 1/0 columns.
Reshape DataChoose Wide to Long or Long to Wide, set the variables, applySwitch between one row per person and one row per measurement.
The tool names above are exactly as they appear on the cards. Some are written slightly differently in the sections below, for example Rename Variable where the card says Rename Variables.

Variable Types

AnalyZ auto-detects variable types. Numeric columns with more than 10 unique values are treated as continuous; those with 10 or fewer unique values are treated as categorical.

Use Variable Types to override this for individual columns, for example, a Likert scale coded 1–5 has more than 10 possible values if recoded but represents categories. Getting types right is important because they control which tests and chart options appear throughout the app.

You can also change the type directly from the Column Inspector in Data View, click a column header, then click Change in the Variable type row. Both routes update the same setting.

Variable & Value Labels

Define variable labels (column descriptions) and value labels (code to text) for any file type. Only numeric-coded columns are listed, so convert text columns with Encode Text Variable first. There are two routes:

Encode Text Variable

Converts a text column (for example Male/Female or Yes/No) into numeric codes. Codes are assigned alphabetically (1, 2, 3 and so on) and the original text is saved as value labels, so the meaning is kept. Review the mapping before applying, because codes cannot be reordered afterwards.

Create Variable

Add a new column to the dataset using one of four methods:

Give the new variable a name, choose its data type, and optionally add a label and value labels before clicking Create.

Recode Variable

Map old values to new values, for example to collapse age groups or reverse-score Likert items. Select the source variable, add rules row by row, and choose the output: Create new variable from recode (the original is kept) or Recode into same variable (the original values are overwritten, so use Undo if you change your mind). Rules can be single values, ranges (1-5 or 18 thru 65), or thresholds (<18, >65). Two catch-all rules, All other non-missing and All missing, let you set a value for everything not matched.

Rename Variable

Rename variables in one of three ways: Assign new names to chosen variables one by one, Add prefix, or Add suffix to several at once. Variable and value labels follow the new names, and the change is reflected everywhere in the app immediately.

Delete Variables / Delete Observations

Delete Variables removes one or more columns. A preview shows what the dataset will look like before you confirm.

Delete Observations removes rows either by row number(s) (for example 3, 10-15) or by condition, a filter you define (e.g. age < 18).

Duplicate Column

Copies a column as-is and inserts it right after the original as name_copy. The label, value labels and variable type carry over. Use it as a backup before transforming or recoding a variable.

Sort Data

Sort rows by one or more variables, each ascending or descending. Earlier rules take priority and later ones break ties; reorder the rules with the arrows. Blank values always sort to the end. Sorting changes row order for the session, so use Undo to revert.

Manage Duplicates

Identifies fully duplicated rows, or duplicates based on a key column you choose. Preview the duplicates, then remove them keeping either the first or last occurrence.

🔗 Merge Datasets

Join a second dataset to the current one using one or more key variables. The key columns can have different names in the two datasets, and you can add several key pairs. Choose a join type:

Join typeBehaviour
InnerKeep only rows that match in both datasets
LeftKeep all rows from the current dataset; add columns from the second where they match
RightKeep all rows from the second dataset; add columns from the current where they match
OuterKeep all rows from both datasets; fill gaps with missing values

📥 Append Dataset

Stack a second dataset on top of the current one (row-wise). Columns present in one dataset but not the other are filled with missing values. Useful for combining data collected in batches (e.g. survey wave 1 and wave 2).

🔀 Split Column

Splits a text column into multiple new columns using a delimiter you specify (comma, space, semicolon, pipe, tab, or any custom string). Each part becomes a separate column named colname_A, colname_B, colname_C, and so on, inserted immediately after the source column.

Use Preview split before applying to verify how the delimiter will divide the values. Useful for columns such as "A B C", "Yes, No, Maybe", or multi-response fields encoded as delimited strings.

Reshape Data

Converts your dataset between wide and long formats, the two fundamental data structures in analysis:

DirectionWhen to useHow it works
Wide → LongRepeated-measures or longitudinal analysis; many value columns represent the same measurement at different times or conditionsSelect ID variables (columns to keep) and value variables (columns to unpivot). Each value variable becomes a row. Uses pd.melt().
Long → WideCreating a summary matrix; pivoting category values into columnsSelect an index column (row identifier), a columns variable (whose unique values become column headers), and a values column. Uses pd.pivot_table() with an aggregation function you choose.
Variable labels and value labels are cleared after reshaping, because column names change entirely. Re-add labels in Data View after the reshape if needed.

Variable & Value Labels

Adds variable labels (a description for each column) and value labels (for example 1 = Male, 2 = Female) to your dataset. Click Download template to get a CSV pre-filled with your variable names, fill in the labels in Excel or Google Sheets, then upload the file. The labels are applied at once and appear in all analysis outputs and charts. You can also add or edit single labels in the manual editor below the upload area.

Value labels only apply to numeric codes. Labels are saved with your session file (.analyz.json), so you do not need to enter them again next time.

Encode Text Variable

Turns a text column, such as Male/Female, Yes/No or region names, into numeric codes. Select the column, review the codes AnalyZ assigns (1, 2, 3 and so on), and click Encode. The text column is replaced by the numeric codes, and value labels are created automatically so the meaning is kept.

The codes cannot be reordered after encoding. Switch on Show value labels in Data View to see the original text. Undo reverses the encoding, but once you save the session the original text is gone. Encode one column at a time.

Duplicate Column

Copies a column. The copy is inserted straight after the original and named with _copy added (or _copy_2 if that name is taken). It keeps the original's variable label, value labels and variable type. This is useful as a safety copy before you recode or transform a column.

Sort Data

Reorders the rows by one or more variables. Choose the first variable and the direction, then click Add variable to add tie-breakers, for example region first, then score descending within each region. Earlier variables in the list take priority, and you can move them up or down. Blank or missing values always sort to the end.

Sorting changes the row order for the whole session. Use Undo to revert it.

↩ Undo

The ↩ Undo button in the sidebar rolls back the last data management action, and ↪ Redo reapplies it. Up to 20 steps of history are maintained. The dataset, variable labels, and value labels are all restored together.

Worked examples

Short, concrete examples of the most-used tools. Your own variable names will differ.

TaskExampleHow
Recode VariableAge in years has values from 15 to 80. You want groups: under 25, 25 to 44, 45 and over.Recode Variable, choose age, define each range and give it a new label, and save into a new variable so the original stays intact.
Recode VariableA Likert item runs 1 (strongly disagree) to 5 (strongly agree) and you want agree versus not agree.Recode 4 and 5 to 1, and 1 to 3 to 0, into a new variable. Add value labels such as 1 = Agree.
Create VariableYou want a total stress score from five items.Create Variable, choose a formula that adds the five columns (or their mean). Check any reverse-coded item first, as in Cronbach's Alpha.
Merge DatasetsHousehold data in one file and individual data in another, sharing a household ID.Merge on the ID column. Check the row count afterwards, since a merge on a non-unique ID can multiply rows.
Append DatasetTwo survey rounds with the same columns.Append the second file below the first. Add a round variable so you can tell them apart, which Difference-in-Differences needs.
Reshape DataOne row per person per time point (long) but a method wants one row per person (wide), or the reverse.Reshape, choose the ID column and the columns that repeat.
Manage DuplicatesThe same respondent was entered twice.Run Manage Duplicates on the ID column, review the flagged rows, then keep the first or the most complete.
Work in a new variable whenever you recode or compute. If a result looks wrong later, the original is still there to check against. Save the session before large changes, and use Undo for the last step.

Descriptive Analysis answers the first questions of any project: Is my data clean? What does a typical respondent look like? How are the answers spread out? It has five analysis cards, arranged in the order most projects use them. Click a card to open it, and use Back to return to the cards. The optional Filter data panel restricts any analysis to a sub-group, for example women aged 18 to 35.

CardIn one lineTypical question
Data QualityA health check of the datasetAre there missing, duplicated or impossible values?
FrequencyCounts and percentages for each answerWhat share of respondents chose each option?
Cross-TabulateHow two categorical questions line upDo men and women answer differently?
TableA ready-made 'Table 1' for reportsHow do programme and comparison groups compare at baseline?
Summary StatisticsAverage, spread and shape of numeric variablesWhat is the typical income, and how much does it vary?
Descriptive results are the foundation of everything else. A quick Frequency or Summary Statistics run on each variable you plan to use catches most mistakes (a gender column with five spellings, ages recorded in months) before they spoil a test.

Data Quality

Data Quality

Before analysisAny dataset
In plain words. A health check for your spreadsheet. It looks for blank answers, repeated records, impossible values, answers that were rounded to favourite numbers, and interviewers whose work looks different from everyone else's. Problems found here are much cheaper to fix now than after you have written up results.
Use it when
  • You have just received or exported a dataset
  • Data collection is still under way and you want to monitor fieldwork
  • A result looks too good, too odd or too uniform
  • You are about to run any other analysis
Your data needs
  • Any loaded dataset
  • For the Enumerator tab: a column that identifies the interviewer
  • For interview duration: start-time and end-time columns
  • For straight-lining: the grid of rating questions to check

Data Quality: how to run it

1
Open the card and click Configure
Choose the settings the later tabs use: the Enumerator ID variable, Interview start time and Interview end time, the scale items to check for straight-lining, and any range rules (for example age must be between 0 and 100).
2
Open the Overview tab and click Run Overview
One row per variable showing valid answers, missing answers, percentage missing, unique values, duplicate values, whether the column is constant, and numeric outliers. The percentage missing is colour-coded: red above 30%, orange from 10% to 30%, green below 10%.
3
Click Find Duplicates on the Duplicate Rows tab, then Run Value Checks on the Value Issues tab
Duplicate Rows lists records that are identical across every column. Value Issues lists numbers outside your valid range and flags digit preference (heaping).
4
Click Run Enumerator Check on the Enumerator Patterns tab
This tab stays greyed out until you pick an Enumerator ID variable in Configure. It compares interviewers on number of interviews, straight-lining rate and median interview duration, and flags those who stand out.
5
Export and fix
Each table downloads as CSV or Excel. Fix problems in Data Management (recode, delete rows, manage duplicates) and run Data Quality again.

Data Quality: reading the results

You seeIt means
% Missing, coloured redMore than 30% of people did not answer. Ask why before using the variable.
Unique values = 1 (constant)Every respondent has the same value, so the variable cannot explain anything.
Outliers (beyond 3 SD)Values far from the average. They may be typing errors or genuine extremes. Look at them before deciding.
Duplicate groupsSets of identical records. Often the same form submitted twice.
Range violationsValues outside the range you set, such as an age of 214.
Heaping detectedToo many answers end in 0 or 5 (or another favoured digit). Respondents are rounding, which blunts precision.
High straight-lining rateMany respondents gave the same answer to every item in a grid.
Short median durationAn interviewer finishes much faster than colleagues. A prompt to review their work.
Worked example

The situation. A household survey of 600 interviews has just come in from five interviewers.

What AnalyZ shows. Overview shows monthly income is 34% missing (red) and age has 5 outliers. Duplicate Rows finds 6 pairs of identical records. Value Issues shows 5 ages above 110 and reports that 41% of ages end in 0 or 5. Enumerator Patterns shows interviewer E07 with a median interview of 9 minutes against 31 minutes for the others, and 72% straight-lining.

How to read it. The 6 duplicate pairs are probably double submissions, so keep one of each. The five impossible ages are probably typing errors, so check the paper forms. The heaping is typical when people do not know their exact age, and you should avoid analysing age in single years. The missing income needs a decision before analysis (see below). E07's interviews deserve a spot check, because a short duration and heavy straight-lining together suggest the questions were not asked in full. These are leads to follow up, not proof of anything.

Data Quality: things to watch for

Data Quality: key terms

Missing value
A blank where an answer should be.
Outlier
A value unusually far from the rest.
Heaping
Rounding of answers to favourite digits.
Straight-lining
Answering every item in a grid identically, often a sign of low effort.
Enumerator
The person who conducts an interview.

Frequency

Frequency

Describes categoriesOne or many variables
In plain words. A frequency table counts how many people gave each answer and turns the counts into percentages. It is the quickest way to see how your respondents are distributed on a question.
Use it when
  • You want to report the breakdown of a categorical question
  • You want to check that value labels and codes look right
  • You need percentages for a report or chart
Your data needs
  • One or more variables, usually categorical
  • Numeric variables with only a few distinct values also work well

Frequency: how to run it

1
Select variables
Choose one or more from the list. Tick Select all variables to run every column at once.
2
Set the options
Include missing values adds a Missing row. Show chart draws a Bar, Horizontal bar or Pie chart beside the table.
3
Click Generate
One table per variable appears. Charts can be downloaded as PNG from the chart menu.

Frequency: reading the results

You seeIt means
ValueThe answer. If a value label exists it is shown instead of the code.
FrequencyHow many respondents gave that answer.
%Share of all respondents, including those who did not answer.
Valid %Share of respondents who did answer. This is the figure most reports use.
Cumulative %Running total of the valid percentages, top to bottom. Useful for ordered answers such as education.
Weighted count and 95% CIAppear when a survey weight is applied. See Survey weights below.
Worked example

The situation. Highest education level was asked of 500 respondents. 20 skipped the question.

EducationFrequency%Valid %Cumulative %
No schooling7214.415.015.0
Primary16833.635.050.0
Secondary15030.031.381.3
Higher9018.018.8100.0
Missing204.0

How to read it. Report the Valid % column: 35.0% finished primary school only, and half of respondents (50.0%) have primary education or less. The plain % column is smaller (33.6%) only because the 20 non-responders are counted in the base. Say which one you used.

Illustrative figures.

Frequency: things to watch for

Frequency is the best way to spot coding problems. If you see 'Male', 'male' and 'M' as separate rows, merge them with Recode in Data Management.

Frequency: key terms

Valid percent
Percent among people who gave an answer.
Cumulative percent
Percent at or below each category.

Cross-Tabulate

Cross-Tabulate

Two categorical variablesIncludes chi-square
In plain words. A cross-tabulation is a grid. One question runs down the side and another across the top, and every cell counts the people who fall into that combination. Percentages turn the counts into comparisons: of the women, what share own a phone, and of the men?
Use it when
  • You want to compare the breakdown of one category across the groups of another
  • You want a quick test of whether two categorical questions are linked
Your data needs
  • Two categorical variables, such as sex and phone ownership
  • A reasonable number of cases in each cell

Cross-Tabulate: how to run it

1
Choose the Row variable and the Column variable
A good habit is to put the group you want to compare (for example sex) in the rows and the outcome (phone ownership) in the columns.
2
Choose the percentages
Row % adds to 100 across each row. Column % adds to 100 down each column. Total % divides by the whole sample. Row % is selected by default.
3
Click Generate
The grid appears with the chi-square test underneath. If a survey weight is applied you can show either respondent counts or weighted counts.

Cross-Tabulate: reading the results

You seeIt means
Cell counts and percentagesRead the percentages along the direction that adds to 100. With groups in rows, read across each row.
Chi-square, dfThe size of the gap between your counts and the counts expected if the two questions were unrelated, and the degrees of freedom.
p-valueBelow 0.05 means the pattern is unlikely to be chance.
Cramér's VStrength of the link from 0 to 1, described in words: very weak (under 0.1), weak (under 0.3), moderate (under 0.5) or strong.
Worked example

The situation. Phone ownership by sex among 400 respondents (180 women and 220 men).

Owns a phoneDoes notTotal
Women126 (70.0%)54 (30.0%)180
Men187 (85.0%)33 (15.0%)220

What AnalyZ shows. Row percentages are shown, so each row adds to 100%. Chi-square = 12.22, df = 1, p-value below 0.001, Cramér's V = 0.17 (weak).

How to read it. Phone ownership is 70% among women and 85% among men, a gap of 15 percentage points. The gap is significant, so it is unlikely to be chance. Cramér's V of 0.17 describes a weak but real link: sex matters, but it is far from the whole story. Reading the column percentages instead would answer a different question (of all phone owners, what share are women?).

Illustrative figures.

Cross-Tabulate: things to watch for

Cross-Tabulate: key terms

Cell
One box in the grid.
Row percent
Percent within each row.
Column percent
Percent within each column.

Table (Table 1)

Table

Report-readyCompares groups
In plain words. This builds the table that opens most research papers and reports: the characteristics of your sample, side by side for each group. Numbers get an average and spread, categories get counts and percentages, and an optional p-value column tells you which rows differ between groups.
Use it when
  • You need a profile of respondents by programme group, region or sex
  • You want to show that two groups were similar at the start (baseline balance)
Your data needs
  • One grouping variable for the columns
  • One or more variables for the rows, numeric or categorical

Table: how to run it

1
Choose the Variable at the top (header variable)
The groups that become columns, for example programme and comparison.
2
Choose the Variables on the side (row variables)
Everything you want described, in the order you want it to appear.
3
Tick the extras you want
A Total N column, cell counts, and p-values. A Table layout preview shows what you are building.
4
Click Generate table and export
Download to Word or Excel for your report.

Table: reading the results

You seeIt means
Continuous row, for example AgeShown as mean (standard deviation) for each group.
Categorical row, for example SexShown as count and percentage, usually percentage within the group column.
Total columnThe whole sample, for reference.
p-value columnA chi-square test for categorical rows and a t-test or ANOVA for numeric rows. With survey weights these become design-based tests.
Worked example

The situation. A training programme compares 200 participants with 200 non-participants before it starts.

Programme (n=200)Comparison (n=200)p-value
Age, mean (SD)31.4 (8.2)32.0 (8.6)0.46
Female, n (%)104 (52.0%)96 (48.0%)0.44
Years of school, mean (SD)8.9 (3.1)8.7 (3.3)0.55
Employed at baseline, n (%)58 (29.0%)81 (40.5%)0.02

How to read it. The two groups look alike on age, sex and schooling, since all the p-values are well above 0.05. They differ on employment: 29.0% of participants were employed against 40.5% of the comparison group. That matters, because a later difference in employment could reflect where each group started. This is the kind of imbalance that matching or a difference-in-differences design is meant to handle.

Illustrative figures.

Table: things to watch for

Table: key terms

Baseline
The situation before a programme starts.
Header variable
The grouping that forms the columns.

Summary Statistics

Summary Statistics

Numeric variablesGroup by option
In plain words. Summary statistics boil a column of numbers down to a few figures: where the middle is, how spread out the values are, and whether the distribution leans to one side.
Use it when
  • You need the typical value and the variation for age, income, scores, weight
  • You want to compare the numbers across groups (for example by region)
  • You are checking whether a variable is skewed before choosing a test
Your data needs
  • One or more numeric variables
  • An optional categorical variable to split the results by

Summary Statistics: how to run it

1
Choose numeric variables
Select any number of continuous variables.
2
Tick the statistics you want
N, Missing, Mean, Median, Mode, SD, Variance, Min, Max, Range, Skewness and Kurtosis. The default set is N, Missing, Mean, Median, SD, Min and Max.
3
Optionally set Group by
Choose a categorical variable to get a separate row for each group.
4
Click Generate
Read the table and the interpretation below it.

Summary Statistics: reading the results

You seeIt means
N and MissingHow many people have a value, and how many do not.
MeanThe average. Sensitive to extreme values.
MedianThe middle value when sorted. Steady even with extremes.
ModeThe most frequent value.
SD and VarianceHow far values typically sit from the mean. SD is in the same units as the data. Variance is SD squared.
Min, Max, RangeThe smallest and largest values, and the gap between them.
SkewnessNear 0 is symmetric. Above 1 means a long tail of high values. Below -1 means a long tail of low values.
KurtosisTail weight. Near 0 looks like a bell. Above 1 means heavy tails and probable outliers. Below -1 means values bunch in the middle.
Worked example

The situation. Monthly household income for 450 households.

NMissingMeanMedianSDMinMaxSkewness
45012412310380403,2002.4

How to read it. The mean (412) is well above the median (310) and skewness is 2.4, meaning a small number of high-income households pull the average up. For a typical household, quote the median. An SD of 380 is almost as large as the mean, which also shows how unequal the incomes are. Because the data are strongly skewed, use rank-based tests such as Mann-Whitney U or Kruskal-Wallis for comparing income across groups, or log the variable first.

Illustrative figures.

Summary Statistics: things to watch for

Summary Statistics: key terms

Standard deviation
The typical distance of values from the mean.
Skewness
Which way the distribution leans.
Kurtosis
How heavy the tails are.

Survey weights

When a survey weight is set, Frequency, Cross-Tabulate, Table and Summary Statistics use it by default. Untick Apply survey weight beside Generate to compare with the unweighted result.

ModuleWhat changes
FrequencyPercentages are weighted, a weighted count is added, and each percentage comes with a 95% confidence interval that reflects the weights, strata and clusters. The Frequency column stays the number of respondents.
Summary StatisticsThe mean, median and measures of spread are weighted. Weighted N, the standard error of the mean and its 95% confidence interval can be added as statistics. N stays the number of respondents.
Cross-TabulatePercentages are weighted. The counts shown can be the number of respondents or the weighted count. The chi-square test is replaced by the Rao-Scott corrected F test, which adjusts for the weights, strata and clusters.
TablePercentages, means and standard deviations are weighted, with the same choice of counts. The p-values come from the Rao-Scott F test for categorical rows and from a design-based test of equal means for continuous rows.

Descriptive results tell you what happened in the people you measured. Inferential tests ask the next question: is this pattern big enough, given how many people I measured, that I should believe it is real, rather than a quirk of who happened to be in my sample? If you are new to this, read Statistics Basics first, in particular the parts on p-values and effect sizes.

Inferential Analysis is organised into five groups. Choose a group, then choose the test, set up your variables and click Run test. Every result ends with a written interpretation, and results can be exported to Word, Excel or CSV, added to your Data Canvas, or sent to Sense Making.

GroupTestsQuestion it answers
Compare MeansIndependent t-test, One-sample t-test, Paired t-test, One-way ANOVA, Two-way ANOVA, Repeated-Measures ANOVADo the averages differ?
Compare PercentagesChi-square test, Z-test for one proportion, Z-test for two proportionsDo the percentages differ?
CorrelationPearson, SpearmanDo two things move together?
RegressionLinear, Logistic, Probit, Ordinal, Multinomial, Poisson, Negative Binomial, Hierarchical, MultilevelWhat predicts the outcome, once other things are held steady?
Non-parametric TestsMann-Whitney U, Wilcoxon Signed-Rank, Kruskal-Wallis, FriedmanThe same as the means tests, for ordinal, skewed or small-sample data
Not sure which test fits? Open Guided Analysis from the top bar. It asks about your goal and your data and recommends a test. The page Choosing an Analysis gives the same guidance as tables.

Before you run any test

Compare Means

These tests compare averages. Pick by how many groups you have and whether they are separate people or the same people measured again.

Independent t-test

Compares 2 averagesSeparate groupsNumeric outcome
In plain words. Asks whether the average of a number differs between two separate groups, for example whether reading scores differ between two teaching methods. The test weighs the size of the gap against how much scores vary inside each group and how many people there are. A big gap, tight groups and a large sample make a convincing difference.
Use it when
  • You are comparing the average of a number between exactly two groups
  • The groups are made of different people (men and women, programme and comparison)
Your data needs
  • A numeric outcome such as score, income or age
  • A grouping variable with exactly 2 values
  • Roughly bell-shaped data in each group, or about 30 or more people per group

Independent t-test: how to run it

1
Choose the Dependent variable (numeric)
The number you want to compare, for example a reading score.
2
Choose the Grouping variable
AnalyZ checks that it has exactly 2 values and tells you if it does not. If you have more categories, merge them with Recode, or use One-way ANOVA.
3
Choose Two-tailed or One-tailed
Leave Two-tailed selected unless you decided before collecting data that only one direction matters. Two-tailed asks whether the groups differ at all, in either direction.
4
Click Run test
Read the group table first, then the test table, then the interpretation.

Independent t-test: reading the results

You seeIt means
N, Mean, SD per groupHow many people, their average and their typical spread. Start here: how far apart are the means, in real units?
tHow many standard errors apart the two means are. Further from 0 means stronger evidence.
dfDegrees of freedom: N minus 2. Needed for the p-value, rarely worth interpreting.
p-value and SigBelow 0.05 (one or more stars) means the gap is unlikely to be chance.
Cohen's dSize of the gap in standard deviations: 0.2 small, 0.5 medium, 0.8 large.
Levene pA check that both groups are equally spread out. If it is below 0.05, the spreads differ and the result deserves caution. In that case also run Mann-Whitney U as a cross-check.
Worked example

The situation. 80 pupils took a reading test after being taught by Method A (40 pupils) or Method B (40 pupils).

GroupNMeanSD
Method A4074.29.6
Method B4068.910.4

What AnalyZ shows. t = 2.37, df = 78, p-value 0.020 (*), Cohen's d = 0.53. Levene p is above 0.05. Interpretation: Method A has a significantly higher mean reading score than Method B at the 5% significance level. The difference is medium in practical terms.

How to read it. Method A pupils scored 5.3 points higher on average. The p-value says this gap is unlikely to be chance, and d = 0.53 says it is a medium-sized gap, about half a standard deviation. Both matter: significance tells you the gap is probably real, and d tells you whether it is worth caring about. A significant result with d = 0.1 would be real but small.

Illustrative figures.

Independent t-test: things to watch for

Independent t-test: key terms

t statistic
The gap between means divided by its uncertainty.
Cohen's d
The gap measured in standard deviations.
Levene's test
A check that the groups are equally scattered.

One-sample t-test

Compares 1 average to a targetNumeric outcome
In plain words. Asks whether the average in your sample differs from a fixed number you choose in advance, such as a standard, a target or a neutral midpoint. For example: is average satisfaction on a 5-point scale different from the neutral midpoint of 3?
Use it when
  • You have a benchmark to compare against
  • You want to test whether an average is different from zero, for example average change
Your data needs
  • One numeric variable
  • A test value you decided before looking at the data

One-sample t-test: how to run it

1
Choose the Variable
The numeric column you want to test.
2
Enter the Test value
The benchmark, for example 3 for a neutral midpoint, or 0 for no change.
3
Choose the tails and click Run test
Two-tailed is the safe default.

One-sample t-test: reading the results

You seeIt means
N, Mean, SD, SESample size, average, spread and the standard error of the mean (how much the average would wobble between samples).
Test valueThe benchmark you entered.
t, df, p-value, SigHow far the mean is from the benchmark in standard errors, and whether that is beyond chance.
95% CI lower and upperA range for the true average. If it does not contain the test value, the result is significant.
Worked example

The situation. 80 respondents rated their satisfaction from 1 to 5. Is the average different from the neutral value of 3?

NMeanSDTest valuetdfp-value95% CI
803.460.9234.4779< 0.0013.26 to 3.67

How to read it. The average of 3.46 is 0.46 points above neutral. The interval 3.26 to 3.67 stays clear of 3, so the result is significant. Satisfaction is modestly positive on average. This says nothing about how many people are satisfied. For that, use Frequency or the Z-test for a proportion.

Illustrative figures.

One-sample t-test: things to watch for

One-sample t-test: key terms

Standard error
How much the sample average is likely to wobble from sample to sample.
Test value
The fixed number you compare against.

Paired t-test

Before and afterSame people twiceNumeric outcome
In plain words. Compares two measurements taken on the same people, such as scores before and after a course. Because each person is compared with themselves, differences between people drop out, which makes the test more sensitive than comparing two separate groups.
Use it when
  • Pre and post scores on the same participants
  • Two measurements per person under two conditions
  • Matched pairs, for example twins or couples
Your data needs
  • Two numeric columns, one row per person, for example pre_score and post_score
  • Rows with a missing value in either column are left out
  • If your data are stacked in rows, reshape to wide format first in Data Management

Paired t-test: how to run it

1
Choose the First variable and the Second variable
The two measurements. The difference is worked out as first minus second, so if you choose the before score first, an improvement appears as a negative difference. Choose the later measurement first if you prefer improvement to be positive.
2
Choose the tails and click Run test
Two-tailed is the safe default.

Paired t-test: reading the results

You seeIt means
Mean and SD of each variableThe two averages. Compare them in real units.
Mean diff and SD diffThe average change per person, and how much that change varies between people.
t, df, p-value, SigWhether the average change is beyond what chance would produce.
95% CI lower and upperA range for the true average change. If it excludes 0, the change is significant.
InterpretationNames which measurement is higher and gives a practical size, from the mean change relative to its spread.
Worked example

The situation. 36 participants took a skills test before and after a training course. Post-test is entered first, so improvement is positive.

VariableNMeanSD
Post-test3661.310.1
Pre-test3656.59.7

What AnalyZ shows. Mean diff = 4.8, SD diff = 8.2, t = 3.51, df = 35, p-value 0.001 (**), 95% CI 2.0 to 7.6. Interpretation: Post-test has a significantly higher mean than Pre-test at the 1% significance level. The difference is medium in practical terms.

How to read it. On average people gained 4.8 points, and the interval 2.0 to 7.6 does not include 0. The SD of the change (8.2) shows that individuals varied widely, and some may have gone down. Without a comparison group, you cannot say the course caused the gain, as people might have improved anyway. For that question see Effect Analysis.

Illustrative figures.

Paired t-test: things to watch for

Paired t-test: key terms

Paired data
Two measurements linked to the same person or matched unit.
Mean difference
The average of each person's change.

One-way ANOVA

Compares 3+ averagesSeparate groupsNumeric outcome
In plain words. Asks whether the average of a number differs across three or more groups at once, for example scores in three districts. It works by comparing how far apart the group averages are against how much people vary inside each group. If the averages are further apart than the inside-group variation would produce on its own, at least one group differs.
Use it when
  • Three or more separate groups
  • You want one overall answer before looking at pairs
Your data needs
  • A numeric outcome
  • A grouping variable with 3 or more values
  • Reasonably similar spread in each group, and at least a handful of people per group

One-way ANOVA: how to run it

1
Choose the Dependent variable (numeric)
The number you are comparing.
2
Choose the Grouping variable
The categories that define the groups.
3
Keep the post-hoc box ticked
Run pairwise post-hoc comparisons (when p < 0.05) is on by default. It lists which pairs of groups differ, but only when the overall test is significant.
4
Click Run test
Read the group table, then the ANOVA table, then the pair-by-pair table.

One-way ANOVA: reading the results

You seeIt means
Group, N, Mean, SDThe averages you are comparing.
FThe ratio of between-group to within-group variation. Larger means stronger evidence of a difference somewhere.
p-value and SigWhether the overall differences are beyond chance. It does not say which groups differ.
η² (eta-squared)The share of all variation explained by group: 0.01 small, 0.06 medium, 0.14 large.
Post-hoc comparisonsOne row per pair of groups, with the difference in means and a p-value. Use this table to see where the difference is.
Worked example

The situation. 105 adults from three districts completed a knowledge test (35 per district).

DistrictNMeanSD
North3558.211.0
Central3564.510.4
South3567.111.6

What AnalyZ shows. F(2, 102) = 6.05, p-value 0.003 (**), η² = 0.11. Post-hoc comparisons: North vs Central, mean diff -6.3, p 0.016 (*). North vs South, mean diff -8.9, p 0.002 (**). Central vs South, mean diff -2.6, p 0.327 (ns).

How to read it. The overall test is significant, and η² = 0.11 says district explains about 11% of the variation in scores, a medium-to-large effect. The pair table shows where the difference sits. North scores lower than both other districts, while Central and South do not differ from each other. Report it as 'North was lower than Central and South', not as 'all three districts differ'.

Illustrative figures.

One-way ANOVA: things to watch for

One-way ANOVA: key terms

Between-group variation
How far group averages sit from the overall average.
Within-group variation
How much people differ inside their own group.
Post-hoc test
A follow-up comparing pairs after the overall test is significant.

Two-way ANOVA

Compares averagesTwo grouping variablesFinds interactions
In plain words. Looks at two grouping variables at once, for example teaching method and sex, and asks three questions: does method matter, does sex matter, and does the effect of one depend on the other? That third question, called an interaction, is often the most interesting finding.
Use it when
  • You have two categorical factors and one numeric outcome
  • You suspect an effect works differently for different groups
Your data needs
  • A numeric outcome
  • Two grouping variables (Factor A and Factor B)
  • Reasonable numbers in every combination of the two, at least 5 per cell and ideally many more

Two-way ANOVA: how to run it

1
Choose the Dependent variable (numeric)
The outcome.
2
Choose Factor A and Factor B
The two categorical variables.
3
Click Run test
The ANOVA table has one row for each factor, one for their interaction, and one for error. Read the interaction row first.

Two-way ANOVA: reading the results

You seeIt means
Factor A and Factor B (main effects)Does the average differ across the categories of that factor, ignoring the other?
A × B (interaction)Does the effect of one factor change across the levels of the other? A significant interaction means it depends.
F, p-value, SigFor each row: the strength of evidence and whether it is beyond chance.
ErrorVariation left over, inside the groups.
Worked example

The situation. 200 pupils took a test, 50 in each combination of teaching method (A or B) and sex.

FemaleMaleAll
Method A80.070.075.0
Method B69.070.069.5
All74.570.0

What AnalyZ shows. Method: F(1, 196) = 18.67, p-value < 0.001. Sex: F(1, 196) = 12.50, p-value 0.001. Method × Sex: F(1, 196) = 18.67, p-value < 0.001.

How to read it. All three rows are significant, but the interaction changes how the other two should be told. The cell means show that Method A helps girls (80 against 69 under Method B) but makes no difference for boys (70 against 70). Saying 'Method A is better' would be misleading. The accurate statement is that Method A works for girls only. When the interaction is significant, describe the cell means rather than the overall method or sex averages.

Illustrative figures.

Two-way ANOVA: things to watch for

Two-way ANOVA: key terms

Main effect
The effect of one factor averaged over the other.
Interaction
When the effect of one factor depends on the level of the other.

Repeated-Measures ANOVA

3+ time pointsSame peopleNumeric outcome
In plain words. Extends the paired t-test to three or more measurements on the same people, such as scores at baseline, midline and endline. It asks whether the average changes across the time points.
Use it when
  • The same people are measured at three or more times or under three or more conditions
  • The outcome is numeric and roughly bell-shaped
Your data needs
  • Three or more numeric columns, one per time point, such as score_baseline, score_midline, score_endline
  • One row per person, with a value in every column
  • Optionally a subject ID column

Repeated-Measures ANOVA: how to run it

1
Choose the Conditions / time points
Select 3 or more numeric columns in time order.
2
Choose a Subject ID column if you have one
It is optional. If left blank, each row is treated as one subject.
3
Click Run test
Read the condition means, the ANOVA table, the sphericity check and the pairwise table.

Repeated-Measures ANOVA: reading the results

You seeIt means
Condition meansThe average at each time point. Look at the pattern first.
F and p-valueWhether the averages differ across time points.
Partial η²How much of the variation is explained by time: 0.01 small, 0.06 medium, 0.14 large.
Mauchly's WA check that the changes between time points vary evenly (sphericity). Close to 1 is fine. Below 0.75, treat the p-value with caution.
Pairwise comparisons (Bonferroni)Each pair of time points compared, with p-values made stricter for the number of comparisons. Shown when the overall result is significant.
Worked example

The situation. 30 participants were tested at baseline, midline and endline.

Time pointMean
Baseline53.6
Midline56.2
Endline61.8

What AnalyZ shows. F(2, 58) = 55.96, p-value < 0.001, partial η² = 0.66. Mauchly's W is above 0.75, so sphericity looks acceptable. All three pairs differ after Bonferroni adjustment.

How to read it. Scores rose steadily, and time explains a large share (66%) of the variation. The pairwise table says each step is a real change, not just the overall baseline-to-endline gain. As with the paired t-test, you cannot tell from this alone whether a programme caused the rise.

Illustrative figures.

Repeated-Measures ANOVA: things to watch for

Repeated-Measures ANOVA: key terms

Sphericity
The assumption that differences between time points vary equally.
Partial eta-squared
The share of variation explained by time once person-to-person differences are set aside.

Compare Percentages

These tests compare shares of people rather than averages. Use them when your variables are categories.

Chi-square test

Two categorical variablesTests for a link
In plain words. The chi-square test asks whether two categorical questions are linked. It counts how many people fall into each combination and compares that with the counts you would see if the questions had nothing to do with each other. The further the real counts are from those 'nothing is going on' counts, the stronger the evidence of a link.
Use it when
  • Both variables are categories (education by employment, region by whether a service was used)
  • You want to know whether groups differ in how they answered
Your data needs
  • Two categorical variables
  • Enough people in each cell. About 5 expected per cell is the minimum.

Chi-square test: how to run it

1
Choose Variable 1 and Variable 2
Both categorical. AnalyZ builds the table of counts for you. Numeric variables with many values need grouping first, with Recode.
2
Click Run test
You get the table of observed counts, a second table of expected counts, the test result, and a warning if any cell has an expected count below 5.

Chi-square test: reading the results

You seeIt means
Observed countsThe real number of people in each combination.
Expected countsHow many you would see in each cell if the two questions were unrelated. Compare with observed to see where the pattern is.
Chi-square (χ²) and dfThe overall size of the gap between observed and expected, with the degrees of freedom.
p-value and SigWhether the gap is beyond chance. Significant means the questions are linked.
Cramér's VStrength of the link from 0 to 1. Roughly: under 0.1 negligible, 0.1 to 0.3 small, 0.3 to 0.5 moderate, above 0.5 strong.
Worked example

The situation. 200 people took part in a job programme (100) or a comparison group (100). Six months later each was recorded as employed full-time, employed part-time, or not employed.

Full-timePart-timeNot employed
Programme (observed)483022
Comparison (observed)274132
Expected in each row37.535.527.0

What AnalyZ shows. Chi-square = 9.44, df = 2, p-value 0.009 (**), Cramér's V = 0.22.

How to read it. Full-time employment is 48% in the programme group and 27% in the comparison group. The expected count for each group was 37.5 people, so the programme group has 10.5 more full-timers than 'nothing is going on' would give. The result is significant, and V = 0.22 is a small to moderate link. Chi-square says the two variables are linked overall. It does not tell you which cell is responsible, so compare observed and expected counts or the row percentages from Cross-Tabulate.

Illustrative figures.

Chi-square test: things to watch for

Chi-square test: key terms

Contingency table
A grid of counts for two categorical variables.
Expected count
The count you would see if the variables were unrelated.
Cramér's V
A 0 to 1 measure of how strongly two categories are linked.

Z-test for one proportion

One percentage vs a targetYes/No outcome
In plain words. Asks whether the share of people with some characteristic differs from a target or expected share. For example: is the share of respondents who improved different from the 50% you hoped for?
Use it when
  • You have a target or benchmark percentage
  • You want to know whether your observed percentage is meaningfully above or below it
Your data needs
  • One categorical variable, usually yes/no
  • A hypothesised proportion between 0 and 1, decided before looking at the data

Z-test for one proportion: how to run it

1
Choose the mode
Select One proportion vs hypothesis.
2
Choose the Variable and the Success value
The success value is the answer you are counting, for example Yes or Improved.
3
Enter the Hypothesised proportion
A number between 0 and 1: 0.5 for 50%, 0.7 for 70%.
4
Click Run test
You get your observed proportion beside the hypothesised one, with z, p-value and a bar chart.

Z-test for one proportion: reading the results

You seeIt means
Observed pThe share in your sample with the success value.
Hypothesised pThe target you entered.
zHow many standard errors the observed share is from the target.
p-value and SigWhether the gap from the target is beyond chance.
Worked example

The situation. 150 participants were assessed and 93 improved. The programme aimed for more than half to improve.

NObserved pHypothesised pzp-value
1500.620.502.940.003

How to read it. 62% improved, which is 12 percentage points above the 50% target, and the gap is unlikely to be chance. Interpretation: The proportion of Improved (62.0%) is significantly different from the expected 50.0% at the 1% significance level. For a statement about how much above 50% the true figure is, use a confidence interval from Frequency.

Illustrative figures.

Z-test for one proportion: things to watch for

Z-test for one proportion: key terms

Proportion
A share expressed as a number from 0 to 1.
Success value
The answer you are counting as a hit.

Z-test for two proportions

Compares 2 percentagesYes/No outcomes
In plain words. Asks whether two percentages are different, for example the share who passed a test in the first round and the share who passed in the second. In AnalyZ the two percentages come from two different variables (columns) in your dataset.
Use it when
  • You want to compare the share with a characteristic across two measures, such as baseline and endline columns
  • You want a direct test of a gap between two percentages
Your data needs
  • Two categorical variables with a success value each
  • Both measured on comparable people

Z-test for two proportions: how to run it

1
Choose the mode
Select Two proportions.
2
Choose Variable 1 and its Success value
For example passed_round1 and the value 1.
3
Choose Variable 2 and its Success value
For example passed_round2 and the value 1.
4
Click Run test
A group table gives N, number of successes and proportion for each variable. The test table gives z, p-value and the difference.

Z-test for two proportions: reading the results

You seeIt means
N, Successes, ProportionEach variable's count and share.
DiffThe first proportion minus the second.
z, p-value, SigWhether the difference is beyond chance.
Worked example

The situation. 54 of 120 households used a service in the first year (45.0%), and 72 of 130 in the second year (55.4%).

NSuccessesProportion
Year 1120540.450
Year 2130720.554

What AnalyZ shows. Diff = -0.104, z = -1.64, p-value 0.101 (ns).

How to read it. Usage was about 10 percentage points higher in year 2, but with these sample sizes the gap is within what chance can produce, so you cannot claim a real rise. A bigger sample might settle it. Say 'no significant difference' rather than 'no difference', because the test cannot prove there is none.

Illustrative figures. The test treats the two groups as independent samples.

Z-test for two proportions: things to watch for

Z-test for two proportions: key terms

Percentage-point difference
The simple gap between two percentages.

Correlation

Correlation measures how closely two numeric variables move together, on a scale from -1 to +1. It does not say that one causes the other.

Value of rReading
+1Perfect upward line: as one rises, the other rises by a steady amount
+0.5Strong positive link
+0.3Moderate positive link
+0.1Weak positive link
0No linear link
-0.3Moderate negative link: as one rises, the other tends to fall
-1Perfect downward line
On this pagePearsonSpearman

Pearson Correlation

Two or more numeric variablesStraight-line link
In plain words. Measures how closely a cloud of points follows a straight line. Pick any number of numeric variables and AnalyZ gives you a table of correlations for every pair, so you can scan for the strongest links.
Use it when
  • Two numeric variables that you expect to rise or fall together
  • A first look at many variables at once
Your data needs
  • Two or more numeric variables
  • A roughly straight-line relationship and no extreme outliers
  • Complete answers on all selected variables. Only people with a value on every selected variable are used.

Pearson Correlation: how to run it

1
Choose the Variables (numeric)
Two or more.
2
Choose display options
Show significance stars, Half matrix to show each pair once, and Colour heatmap to shade the table by strength.
3
Click Run test
A correlation matrix appears, with a significance level for each pair.

Pearson Correlation: reading the results

You seeIt means
rThe correlation for each pair, from -1 to +1. The sign gives direction and the size gives strength.
Stars* p-value below .05, ** below .01, *** below .001. No stars means the link is not distinguishable from chance.
Heatmap coloursDarker cells are stronger. One colour is positive and the other negative.
r squared (mentally)Square r to get the share of variation shared. r = 0.5 means about 25% is shared.
Worked example

The situation. 80 learners recorded their weekly study hours and their exam score.

Study hoursExam score
Study hours1.0000.639 ***
Exam score0.639 ***1.000

How to read it. r = 0.64 is a strong positive link: learners who studied more tended to score higher, and study hours account for about 41% of the variation in scores (0.64 squared). The three stars say this is very unlikely to be chance. It does not prove that studying caused the higher scores, since motivation or prior ability could drive both.

Illustrative figures.

Pearson Correlation: things to watch for

Pearson Correlation: key terms

r
The Pearson correlation coefficient.
Linear
Following a straight line.

Spearman Correlation

RanksOrdinal or skewed data
In plain words. Spearman works like Pearson, but on ranks instead of raw values. Each variable is first turned into positions (1st, 2nd, 3rd and so on), and the correlation is computed on those. This makes it steady when there are outliers, when the data are skewed, or when the variables are ordinal such as Likert scales.
Use it when
  • Ordinal variables such as ratings or education level
  • Skewed data, or data with outliers
  • The pattern rises or falls steadily but not along a perfectly straight line
Your data needs
  • Two or more variables, numeric or ordered categories
  • Only cases with a value on every selected variable are used

Spearman Correlation: how to run it

1
Choose the Variables
Two or more. Unlike Pearson, ordinal variables are accepted.
2
Choose display options and click Run test
The same display options as Pearson.

Spearman Correlation: reading the results

You seeIt means
ρ (rho)The Spearman correlation, read exactly like r.
StarsSignificance, as with Pearson.
Worked example

The situation. The same 80 learners also rated their motivation from 1 to 5.

What AnalyZ shows. Spearman ρ between motivation and exam score = 0.43 (***).

How to read it. Higher motivation goes with higher scores, a moderate link. Because motivation is a 1-to-5 rating, Spearman is the right choice. Pearson would treat the gap between 1 and 2 as equal to the gap between 4 and 5, which ratings do not guarantee.

Illustrative figures.

Spearman Correlation: things to watch for

Spearman Correlation: key terms

Rank
A value's position when all values are sorted.
Monotone
Always rising, or always falling, but not necessarily in a straight line.

Regression

Regression answers the question what goes with the outcome? when several things could matter at once. It looks for the pattern that best links your outcome to a set of predictors, and reports how much the outcome changes for each predictor while the other predictors are held steady. That last part is the strength of regression: it separates the effect of schooling from the effect of age, for instance, even when older people tend to have less schooling.

Four ideas that apply to every regression

IdeaIn plain words
Outcome and predictorsThe outcome (dependent variable) is what you want to explain. The predictors (independent variables) are the things you think are related to it.
The coefficientOne number per predictor: how much the outcome changes when that predictor goes up by one unit, with everything else held constant. It may be shown as B, an odds ratio or a rate ratio, depending on the model.
Categorical predictors and the reference categoryA category variable is turned into a set of yes/no comparisons against one reference category, which AnalyZ chooses as the first category by default. Use Change reference category if another group makes a more natural comparison (for example the largest group, or the untreated group). Every coefficient then reads 'compared with the reference'.
Fit and significanceEach predictor has its own p-value and 95% confidence interval. The model as a whole has a fit measure (R-squared for a numeric outcome, pseudo R-squared for others) saying how much of the outcome the predictors explain.
Pick the regression by the type of outcome. A numeric outcome uses Linear, yes/no uses Logistic, ordered categories use Ordinal, unordered categories use Multinomial, and counts use Poisson or Negative Binomial. People nested in groups need Multilevel.
Regression shows association, not cause. Holding other measured variables constant helps, but anything you did not measure can still be distorting the result. For cause-and-effect questions, see Effect Analysis.

Linear Regression

Numeric outcomeOne or more predictors
In plain words. Finds the straight-line pattern that best predicts a number from other variables. If income goes up by about 50 for every extra year of schooling, linear regression estimates that 50 and tells you how sure it is, after allowing for age, sex and anything else you include.
Use it when
  • Your outcome is a number: income, score, weight, days to recovery
  • You want the effect of each predictor with the others held constant
Your data needs
  • A numeric outcome
  • One or more predictors, numeric or categorical
  • A reasonable sample: at least 10 to 20 people for every predictor
  • A broadly straight-line relationship and no extreme outliers

Linear Regression: how to run it

1
Choose the Dependent variable (numeric)
The outcome you want to explain.
2
Choose the Independent variables
Add every predictor you want to include. Categorical ones are converted to comparisons against a reference category.
3
Change reference categories if needed
Use Change reference category beneath the predictor list.
4
Click Run regression
Read the model summary first, then the coefficient table, then the interpretation.

Linear Regression: reading the results

You seeIt means
NPeople used. Anyone missing on any variable is left out.
R² and Adj R²Share of the variation in the outcome explained by all the predictors together. Adjusted R² penalises predictors that add nothing. Prefer it when comparing models of different sizes.
F and its p-valueWhether the predictors as a group explain anything.
BThe coefficient: change in the outcome for one more unit of the predictor, others held constant. For a category it is the gap from the reference category.
SEHow uncertain B is.
p-value and SigWhether that predictor's effect is distinguishable from zero.
95% CIThe plausible range for B. If it includes 0, the predictor is not significant.
AICA fit score used to compare models on the same data. Lower is better. It has no meaning alone.
Worked example

The situation. 200 adults. Outcome: monthly income. Predictors: age in years, years of schooling, and sex (male is the reference).

PredictorBSEp-value95% CI
(Intercept)373.6108.10.001160.4 to 586.9
Age19.82.2< 0.00115.5 to 24.1
Years of schooling51.35.7< 0.00140.0 to 62.6
Female (vs Male)-95.037.00.011-167.9 to -22.1

What AnalyZ shows. R² = 0.44, Adj R² = 0.43, F(3, 196) = 51.79, p-value < 0.001.

How to read it. Each extra year of schooling goes with about 51 more in monthly income, and each extra year of age with about 20 more, holding the other predictors constant. Women earn about 95 less than men of the same age and schooling. All three intervals exclude zero. The model explains 44% of the variation in income, so more than half is down to things not in the model. The intercept is the predicted income for a man aged 0 with no schooling. It is a mathematical anchor, not a real person, so ignore it.

Illustrative figures.

Linear Regression: things to watch for

Linear Regression: key terms

B
The unstandardised coefficient, in the outcome's own units.
R²
The proportion of variation explained.
Reference category
The comparison group for a categorical predictor.
Holding constant
Comparing people who are alike on the other predictors.

Logistic Regression

Yes/No outcomeReports odds ratios
In plain words. Predicts a yes/no outcome, such as whether someone attended, recovered or dropped out, from other variables. Instead of a straight line it models the odds of the outcome, so predictions always stay between 0% and 100%. AnalyZ reports odds ratios, which say how many times higher or lower the odds are for each predictor.
Use it when
  • The outcome has exactly two values: yes/no, 0/1, pass/fail
  • You want to know which factors are linked to the outcome and how strongly
Your data needs
  • A binary outcome with a decent number of cases in each category. A rough guide is at least 10 of the rarer outcome per predictor.
  • One or more predictors, numeric or categorical

Logistic Regression: how to run it

1
Choose the Dependent variable
It must have exactly two values.
2
Choose which value is the event of interest
The selector Which value is the event of interest? decides which answer the odds ratios are about. Choose Yes, not No, if you want to explain who does it.
3
Choose the Independent variables and reference categories
Predictors can be numeric or categorical.
4
Click Run regression
Use the OR / Coefficient (B) switch above the table to move between odds ratios and log-odds.

Logistic Regression: reading the results

You seeIt means
OR (odds ratio)1 means no effect. Above 1: the odds of the event are higher. Below 1: lower. An OR of 1.5 is 50% higher odds, and 0.80 is 20% lower odds, for each one-unit increase.
95% CI on the ORIf it includes 1, the predictor is not significant.
p-value, Sig, Wald zWhether that predictor's effect is beyond chance.
LR chi2 and Prob > chi2Whether the predictors as a group improve on a model with none.
Pseudo R²A rough fit measure. Values look small compared with linear R², and 0.2 to 0.4 is already considered a good fit.
AIC and Log-likelihoodModel-comparison scores. Lower AIC is better.
Accuracy %The share of cases the model classifies correctly at a 50% cut-off. Compare it with simply guessing the commoner outcome.
ConvergedIf No, the estimates are not reliable. See the warnings below.
Worked example

The situation. 400 people were asked whether they attended a clinic in the last year (55% did). Predictors: distance to the clinic in km, and whether the household owns a means of transport.

PredictorOR95% CIp-value
Distance to clinic (per km)0.800.73 to 0.88< 0.001
Owns transport (vs does not)1.490.99 to 2.250.055

What AnalyZ shows. LR chi2 = 29.78, df = 2, p-value < 0.001. Pseudo R² = 0.05. Accuracy = 60%.

How to read it. Each extra kilometre lowers the odds of attending by 20%, a clear and significant effect. Owning transport raises the odds by 49%, but the interval (0.99 to 2.25) just touches 1, so there is no firm evidence of an effect in this sample. The pseudo R² is small, which is normal when behaviour has many causes. The accuracy of 60% is only modestly better than the 55% you would get by always predicting attendance, so use the model to understand the factors rather than to predict individuals.

Illustrative figures.

Logistic Regression: things to watch for

Logistic Regression: key terms

Odds
Chance of the event divided by chance of no event. A 75% chance is odds of 3 to 1.
Odds ratio
How the odds change per unit of the predictor.
Log-odds
The raw scale the model works on. The B column.

Probit Regression

Yes/No outcomeAlternative to logistic
In plain words. Probit regression answers the same question as logistic regression, using a slightly different mathematical shape (the normal curve rather than the logistic curve). The two nearly always agree on which predictors matter.
Use it when
  • Your field or supervisor expects probit, as is common in economics and some health research
  • You want to check that a logistic result does not depend on the model shape
Your data needs
  • The same as logistic: a yes/no outcome and one or more predictors

Probit Regression: how to run it

1
Set up exactly as for Logistic Regression
Choose the binary outcome, the event of interest, the predictors and any reference categories.
2
Click Run regression
The table gives coefficients on the probit scale. There is no odds ratio.

Probit Regression: reading the results

You seeIt means
BThe change in the underlying 'probit score' per unit of the predictor. The sign tells you direction. The size has no everyday meaning.
p-value, Sig, 95% CIRead exactly as in other models.
Model fit rowsLog-likelihood, LR chi2, pseudo R² and AIC, as in logistic regression.
Worked example

The situation. The clinic attendance data from the logistic example, run as a probit model.

What AnalyZ shows. Distance: B = -0.13, p-value < 0.001. Owns transport: B = 0.24, p-value 0.06.

How to read it. The signs and significance agree with the logistic result: distance lowers attendance and transport may raise it. You cannot say 'by how much' in plain terms from a probit coefficient. If you need a size that readers will understand, report the logistic odds ratios alongside or instead.

Illustrative figures.

Probit Regression: things to watch for

Probit Regression: key terms

Probit
A model for yes/no outcomes using the normal curve.

Ordinal Regression

Ordered categoriesReports odds ratios
In plain words. For outcomes with a natural order, such as satisfaction from low to high, or mild to severe. It asks which predictors push people towards the higher end of the scale, and respects the order without pretending the gaps between steps are equal.
Use it when
  • The outcome is a Likert item or a severity or grade rating
  • Treating the scale as a plain number (linear regression) seems doubtful
Your data needs
  • An ordered outcome with 3 or more categories
  • One or more predictors
  • Enough people in each category

Ordinal Regression: how to run it

1
Choose the Dependent variable
The ordered outcome. Make sure its codes run in the right order, for example 1 = low to 3 = high.
2
Choose the Independent variables
And any reference categories.
3
Click Run regression
Use the OR / Coefficient (B) switch to move between odds ratios and log-odds.

Ordinal Regression: reading the results

You seeIt means
OR for each predictorThe change in the odds of being in a higher category rather than a lower one, per unit of the predictor. Above 1 pushes people up the scale. Below 1 pushes them down.
Threshold rows (/cut1, /cut2 …)Where the scale divides between categories. They are mechanical and rarely reported.
p-value, 95% CI, SigAs in other models.
LR chi2, pseudo R², AICOverall fit.
Worked example

The situation. 500 patients rated service as Low, Medium or High. Predictors: waiting time in 10-minute blocks, and whether the patient has insurance.

PredictorOR95% CIp-value
Waiting time (per 10 min)0.780.70 to 0.87< 0.001
Insured (vs not insured)1.901.30 to 2.800.001

How to read it. Each extra 10 minutes of waiting cuts the odds of a higher rating by 22%. Insured patients have 90% higher odds of rating the service higher. The word 'higher' means any step up, whether Low to Medium or Medium to High. One odds ratio summarises all the steps.

Illustrative figures.

Ordinal Regression: things to watch for

Ordinal Regression: key terms

Proportional odds
The assumption that a predictor's effect is the same across every step of the scale.
Threshold
The cut-points between adjacent categories.

Multinomial Regression

3+ unordered categoriesOne result set per category
In plain words. For outcomes with three or more categories that have no order, such as which programme someone chose. It compares each category with one reference category and shows what predicts being in each, rather than the reference.
Use it when
  • The outcome is a choice or type: programme chosen, main occupation, source of water
  • Categories cannot be ranked
Your data needs
  • An outcome with 3 or more unordered categories
  • One or more predictors
  • A decent number of people in every category

Multinomial Regression: how to run it

1
Choose the Dependent variable
The multi-category outcome.
2
Choose the Reference category (outcome)
Pick the category that makes the most natural comparison: the commonest, or the 'default' choice. Every result is read against it.
3
Choose the Independent variables and click Run regression
Results are shown separately for each non-reference category.

Multinomial Regression: reading the results

You seeIt means
Outcome blocksOne block of rows per non-reference category. A block answers: what predicts choosing this category over the reference?
RRR (relative risk ratio)Above 1: the predictor raises the chance of choosing this category relative to the reference. Below 1: lowers it. Use the switch to see the raw coefficient B instead.
p-value, 95% CI, SigAs in other models.
LR chi2, pseudo R², AICOverall fit.
Worked example

The situation. 600 trainees chose Farming, Vocational or Business training. Farming is the reference. Predictors: age and sex.

Outcome vs FarmingPredictorRRRp-value
VocationalAge (per year)0.94< 0.001
VocationalFemale (vs male)1.100.55
BusinessAge (per year)1.010.62
BusinessFemale (vs male)2.100.002

How to read it. Older trainees are less likely to choose Vocational over Farming (6% lower per year of age). Women are more than twice as likely as men to choose Business over Farming. Sex makes no clear difference to choosing Vocational over Farming, and age makes no clear difference to choosing Business. Notice that every statement is a comparison with Farming. Nothing here says anything about Vocational against Business.

Illustrative figures.

Multinomial Regression: things to watch for

Multinomial Regression: key terms

Reference category
The outcome everything else is compared with.
Relative risk ratio
How the chance of one outcome relative to the reference changes per unit of the predictor.

Poisson Regression

Count outcomeReports rate ratios
In plain words. For outcomes that are counts of events, such as the number of clinic visits or incidents: whole numbers that start at zero. It models how the rate at which events happen changes with the predictors, and reports incidence rate ratios (IRR).
Use it when
  • The outcome counts events
  • Counts are not wildly spread out (the variance is close to the average)
Your data needs
  • A count outcome (0, 1, 2, …) with no negative or decimal values
  • One or more predictors

Poisson Regression: how to run it

1
Choose the Dependent variable (count)
A non-negative whole-number variable.
2
Choose the Independent variables
Categorical predictors are compared against the first category unless you change the reference.
3
Click Run regression
Check the overdispersion line first. Use the IRR / Coefficient (B) switch to see log-scale results.

Poisson Regression: reading the results

You seeIt means
IRR1 means no effect. 1.30 means the count is 30% higher for each extra unit (or for that category compared with the reference). 0.80 means 20% lower.
95% CI, p-value, SigAs in other models.
Dispersion (Pearson chi-square / N)Close to 1 is fine for Poisson. Above about 1.5 the counts are more spread than Poisson allows, and the p-values will look too good. Switch to Negative Binomial.
Worked example

The situation. 300 households reported clinic visits in a year (average 2.9, but a variance of 8.3). Predictors: an asset score (per standard deviation) and whether the household is in a health scheme.

PredictorIRR95% CIp-value
Asset score (per SD)1.331.24 to 1.44< 0.001
In health scheme (vs not)1.301.14 to 1.49< 0.001

What AnalyZ shows. Dispersion = 2.56.

How to read it. The IRRs suggest each standard deviation of assets goes with 33% more visits and scheme members have 30% more. But the dispersion of 2.56 is well above 1.5, so the variance is far larger than the average and these intervals are too narrow. Move to Negative Binomial and trust that result instead.

Illustrative figures.

Poisson Regression: things to watch for

Poisson Regression: key terms

IRR
Incidence rate ratio: how the average count changes per unit of a predictor.
Overdispersion
More variation in counts than Poisson allows.

Negative Binomial Regression

Count outcomeHandles overdispersion
In plain words. The same idea as Poisson regression, but it allows the counts to be more spread out than Poisson permits, which is common in real data where a few people have very high counts. It adds a single extra number, Alpha, to measure how much extra spread there is.
Use it when
  • A Poisson run flagged overdispersion
  • Your counts are very uneven, with many small values and a few big ones
Your data needs
  • The same as Poisson

Negative Binomial Regression: how to run it

1
Set up exactly as for Poisson
Choose the count outcome and predictors.
2
Click Run regression
Read IRRs as before, and note Alpha.

Negative Binomial Regression: reading the results

You seeIt means
IRR, 95% CI, p-valueRead as in Poisson, but with intervals that honestly reflect the extra variation.
AlphaClose to 0 means Poisson would have done. Larger values mean more overdispersion.
Worked example

The situation. The same 300 households.

PredictorIRR95% CIp-value
Asset score (per SD)1.361.20 to 1.53< 0.001
In health scheme (vs not)1.331.07 to 1.640.010

What AnalyZ shows. Alpha = 0.53.

How to read it. The IRRs are similar to Poisson, but the intervals are wider. The health scheme effect (1.07 to 1.64) is now much less certain, and its p-value rose from below 0.001 to 0.010. This is the honest picture. The Poisson result overstated how sure we were. Alpha of 0.53 confirms substantial overdispersion.

Illustrative figures.

Negative Binomial Regression: things to watch for

Negative Binomial Regression: key terms

Alpha
The extra spread in counts beyond Poisson.

Hierarchical Regression

Numeric outcomeAdds predictors in blocks
In plain words. Builds a linear regression in stages. You enter predictors in groups called blocks, usually background characteristics first and the variables you care about next, and see how much each block adds to the explanation. The key output is the change in R² at each step.
Use it when
  • You want to know whether a set of variables adds anything beyond what you already account for
  • You have a theory about the order in which predictors matter
Your data needs
  • A numeric outcome
  • Two or more blocks of predictors

Hierarchical Regression: how to run it

1
Choose the Dependent variable
The numeric outcome.
2
Define blocks
Add Block 1, usually control variables such as age and sex. Click + Add Block for the next group, such as programme or attitudes.
3
Click Run regression
You get a model for each block, the change statistics (ΔR², ΔF and its significance), and a combined interpretation.

Hierarchical Regression: reading the results

You seeIt means
R² for each blockThe explained share at each stage.
ΔR²How much explanatory power the new block added.
ΔF and its p-valueWhether the added power is more than chance would give.
Block coefficientsThe predictor effects in each stage, which show how effects shift as variables are added.
Worked example

The situation. The income model, built in two blocks: Block 1 = age and sex, Block 2 adds years of schooling.

ModelR²ΔR²ΔFp-value
Block 1: age, sex0.213
Block 2: + schooling0.4420.22980.4< 0.001

How to read it. Age and sex alone explain 21% of the variation in income. Adding schooling lifts this to 44%, so schooling adds 23 percentage points, a substantial and significant improvement. If schooling had added 1% the verdict would be that it does not matter much beyond age and sex.

Illustrative figures.

Hierarchical Regression: things to watch for

Hierarchical Regression: key terms

Block
A group of predictors entered together.
ΔR²
The extra variation explained by a block.

Multilevel Regression

Grouped dataContinuous, binary or count outcomes2 to 4 levels
In plain words. Ordinary regression assumes every person is independent. But pupils in the same school, patients in the same clinic or households in the same village tend to resemble each other. Multilevel regression allows for this by giving each group its own starting level (a random intercept), and reports how much of the variation sits between groups through the ICC.
Use it when
  • People are nested in schools, clinics, villages or districts
  • You have repeated measurements on the same people
  • You want group-level and individual-level predictors in one model
Your data needs
  • An outcome that is numeric, binary or a count
  • A variable that says which group each case belongs to, for each level
  • Several groups. Fewer than 5 groups give unreliable group-level estimates, and 20 or more are comfortable.

Multilevel Regression: how to run it

1
Choose the outcome type and model type
Continuous, Binary or Count. Then Random Intercept (groups differ in baseline, with the same predictor effects, which is right for most surveys) or Random Slope (continuous outcomes only: one predictor also varies by group).
2
Set the number of levels and grouping variables
Choose 2, 3 or 4 levels and a grouping variable for each. A hierarchy check confirms whether the units are truly nested.
3
Select the outcome and the predictors
For a binary outcome, say which value is the event. Predictors can be about individuals or about groups.
4
Click Run model
Read the ICC and variance components first, then the fixed effects. The ICC verification panel compares the model-based ICC with a simple one from group averages.

Multilevel Regression: reading the results

You seeIt means
Fixed effectsThe average effect of each predictor across all groups: B for continuous outcomes, odds ratio for binary, rate ratio for counts. Read them like ordinary regression.
Variance componentsHow the unexplained variation splits between groups and individuals.
ICCThe share of variation that lies between groups. Above about 0.05 means grouping matters and an ordinary regression would be misleading.
AIC, BIC, log-likelihoodFit scores for comparing models.
Random intercept BLUPsEach group's own baseline relative to the average, after accounting for the predictors.
Worked example

The situation. 1,200 pupils in 40 schools took a maths test. The predictor is weekly study hours.

PartEstimate
Fixed effect: study hoursB = 2.3, SE 0.4, p-value < 0.001
Variance between schools24.6
Variance between pupils within schools112.1
ICC0.18

How to read it. Each extra weekly study hour goes with 2.3 more marks on average. The ICC of 0.18 says 18% of the variation in marks lies between schools and 82% between pupils in the same school, so schools matter and a model that ignored them would give intervals that are too narrow. Without the multilevel model you would be treating 1,200 pupils as 1,200 independent pieces of evidence, when they behave partly like 40 clusters.

Illustrative figures.

Multilevel Regression: things to watch for

Multilevel Regression: key terms

ICC
Intraclass correlation: the share of variation between groups.
Random intercept
A different starting level for each group.
Fixed effect
The average effect of a predictor across all groups.
Nested
Each lower-level unit belongs to exactly one higher-level unit.

Non-parametric Tests

Non-parametric tests replace the raw values with their ranks (1st, 2nd, 3rd and so on) before comparing groups. That makes them steady against skewed data, outliers and ordinal answers. Each one is the counterpart of a test in Compare Means, and the cost is slightly less power when the data really are bell-shaped.

Instead ofUseWhen
Independent t-testMann-Whitney UTwo separate groups
Paired t-testWilcoxon Signed-RankTwo measurements on the same people
One-way ANOVAKruskal-WallisThree or more separate groups
Repeated-Measures ANOVAFriedman TestThree or more measurements on the same people
Report medians next to these tests. They compare the rank positions of whole groups, and the median is the plain-language way to show the typical value. With survey weights, the tests rank values by their estimated position in the population and report t or F in place of U, H or chi-square (see Survey weights below).

Mann-Whitney U

Compares 2 groupsRanksOrdinal or skewed outcome
In plain words. Pools everyone from both groups, ranks them from lowest to highest, and checks whether one group's people tend to sit higher in the ranking. In practical terms it asks: if I pick one person at random from each group, is one group's person usually higher?
Use it when
  • You are comparing two separate groups on an ordinal rating or a skewed number
  • Samples are small and normality is doubtful
Your data needs
  • A numeric or ordinal variable
  • A grouping variable with exactly two groups

Mann-Whitney U: how to run it

1
Choose the Numeric variable
The rating or score.
2
Choose the Grouping variable (2 groups)
Exactly two values.
3
Click Run test
Medians and mean ranks for each group appear, then the test.

Mann-Whitney U: reading the results

You seeIt means
Median and mean rank per groupThe median is the typical value. The mean rank shows which group sits higher overall. Higher mean rank means higher scores.
UThe test statistic. Smaller means the groups overlap less.
p-value and SigWhether the difference in rank positions is beyond chance.
Effect size r0.1 small, 0.3 medium, 0.5 large.
Worked example

The situation. 60 people rated their satisfaction from 1 to 5 after a service delivered either online (30) or in person (30).

GroupNMedianMean rank
Online30325.8
In person30435.2

What AnalyZ shows. U = 308, p-value 0.030 (*), effect size r = 0.28.

How to read it. People served in person gave higher ratings, with a median of 4 against 3 and a higher mean rank (35.2 against 25.8). The difference is significant, and r = 0.28 is a small-to-medium effect. A t-test would have averaged the 1-to-5 codes, which assumes the steps are equal. Mann-Whitney does not.

Illustrative figures.

Mann-Whitney U: things to watch for

Mann-Whitney U: key terms

Rank
A value's position in the sorted list of all values.
Mean rank
The average rank of a group's members.

Wilcoxon Signed-Rank

Before and afterRanksSame people twice
In plain words. The rank-based version of the paired t-test. It takes each person's change, ranks the size of the changes, and checks whether ups outweigh downs consistently.
Use it when
  • Paired measurements that are ordinal, skewed, or from a small sample
  • The paired t-test's bell-shape assumption looks doubtful
Your data needs
  • Two numeric or ordinal columns for the same people
  • Pairs with a missing value are dropped automatically

Wilcoxon Signed-Rank: how to run it

1
Choose the First variable and the Second variable
For example pre-test and post-test.
2
Click Run test
Medians and the test appear.

Wilcoxon Signed-Rank: reading the results

You seeIt means
MediansThe typical score at each time.
WThe test statistic: the smaller of the sums of ranks for increases and decreases.
p-value and SigWhether the typical change is beyond chance.
Effect size r0.1 small, 0.3 medium, 0.5 large.
Worked example

The situation. 28 participants were scored before and after a short course.

TimeMedian
Before50
After55

What AnalyZ shows. W = 19, p-value < 0.001 (***), effect size r about 0.8.

How to read it. The median rose from 50 to 55, and nearly everyone improved: a small W means the 'down' changes were few and small. The effect is large. Unlike a t-test, a single huge change by one person would not have swung the result.

Illustrative figures.

Wilcoxon Signed-Rank: things to watch for

Wilcoxon Signed-Rank: key terms

Signed rank
The rank of a change's size with its direction (up or down) attached.

Kruskal-Wallis

Compares 3+ groupsRanks
In plain words. The rank-based version of one-way ANOVA. It ranks everyone together and asks whether the groups' average rank positions differ more than chance would allow.
Use it when
  • Three or more separate groups on an ordinal or skewed outcome
  • Small or uneven groups where ANOVA's assumptions are doubtful
Your data needs
  • A numeric or ordinal variable
  • A grouping variable with 3 or more groups

Kruskal-Wallis: how to run it

1
Choose the Numeric variable and the Grouping variable (3+ groups)
At least three groups. For two, use Mann-Whitney U.
2
Click Run test
Group medians and ranks, the overall test, and pair-by-pair comparisons appear.

Kruskal-Wallis: reading the results

You seeIt means
Median and mean rank per groupThe typical value and relative position of each group.
HThe test statistic. Larger means the groups are further apart.
p-value and η²Whether the groups differ, and how much of the ranking is explained: 0.01 small, 0.06 medium, 0.14 large.
Post-hoc comparisons (Dunn, Bonferroni-adjusted)Each pair of groups compared, with p-values made stricter for the number of pairs. Use this table to find which groups differ.
Worked example

The situation. 75 workers in three age bands rated their job stress (25 per band).

BandNMedian stress score
Under 302549
30 to 452554
Over 452567

What AnalyZ shows. H = 29.2, p-value < 0.001 (***), η² about 0.38. In the Dunn table, Over 45 differs from both other bands. Under 30 and 30 to 45 do not differ from each other.

How to read it. Stress is clearly higher among workers over 45, and this stands out against the other two groups, which are similar. The effect is large. A significant H alone would only say 'somewhere there is a difference', so the pair table is where the finding is.

Illustrative figures.

Kruskal-Wallis: things to watch for

Kruskal-Wallis: key terms

Dunn test
Pairwise comparisons that follow a Kruskal-Wallis test.
Bonferroni
A way of making p-values stricter when many pairs are compared.

Friedman Test

3+ measurementsSame peopleRanks
In plain words. The rank-based version of repeated-measures ANOVA. For each person it ranks their measurements, then checks whether certain time points consistently rank higher than others across people.
Use it when
  • Three or more measurements on the same people, with ordinal or skewed outcomes
  • Ratings of several items by the same respondents
Your data needs
  • Three or more numeric or ordinal columns
  • One row per person, with every measurement present

Friedman Test: how to run it

1
Choose 3 or more variables
The repeated measurements, in logical order.
2
Click Run test
Medians per condition, the test and pairwise comparisons appear.

Friedman Test: reading the results

You seeIt means
Median per conditionThe typical value at each time or condition.
Chi-square and p-valueWhether the conditions differ.
Kendall's WAgreement across people, from 0 (none) to 1 (everyone ranks the conditions the same way). It is the effect size.
Pairwise Wilcoxon (Bonferroni)Each pair of conditions compared. Pairs marked ns still show their adjusted p-value, worth checking when borderline.
Worked example

The situation. 30 people were scored at baseline, midline and endline.

What AnalyZ shows. Chi-square = 41.3, p-value < 0.001 (***), Kendall's W = 0.69.

How to read it. W = 0.69 means most people showed the same ordering, with baseline lowest and endline highest. The pairwise table then tells you which steps are real. Use this instead of repeated-measures ANOVA when the scores are ordinal or skewed.

Illustrative figures.

Friedman Test: things to watch for

Friedman Test: key terms

Kendall's W
The degree of agreement between people in how they rank the conditions.

Survey weights

Every test in Inferential Analysis can use the survey weight set in Data View. The Apply survey weight option beside Generate is ticked by default. The tests take the survey design into account (weights, strata and clusters), so p-values and confidence intervals reflect how the sample was drawn. A filter is treated as a subgroup of the full sample.

TestsWhat changes
t-testsMeans are weighted, and the standard error and degrees of freedom come from the survey design. The independent t-test also shows the mean difference with its 95% interval and no longer shows the Levene test, because it does not assume equal variances.
ANOVAThe F test is design-based. One-way and two-way ANOVA show weighted sums of squares and eta squared, and pairwise comparisons are Bonferroni adjusted. Repeated-measures ANOVA uses a joint test that needs no sphericity assumption.
Chi-square testReplaced by the Rao-Scott corrected F test, with weighted frequencies.
Z-test for ProportionThe proportion is weighted. The test uses a t statistic with the design degrees of freedom in place of z.
Pearson and Spearman correlationCorrelations are weighted and the test of no correlation uses the survey design. Spearman is worked out on population ranks estimated from the weights.
Mann-Whitney U and Kruskal-WallisValues are ranked by their estimated position in the population. The tests compare the groups' average rank scores with the survey design and report t or F instead of U or H. This follows the approach of Lumley and Scott (2013).
Wilcoxon Signed-Rank and FriedmanBuilt on the same principle. Wilcoxon scores each difference by its sign and the estimated population rank of its size, and Friedman compares weighted average ranks within persons. Both use the survey design and report t or F.
Regression modelsCoefficients are estimated with the weights, and the standard errors, confidence intervals and p-values come from the survey design. A design-based F test replaces the likelihood-ratio test, adjusted R-squared, AIC and BIC are not shown, and Weighted N appears next to N.
Multilevel RegressionThe survey weight is used within groups, and groups are treated as equally likely to be selected, as in Stata with a single weight. With two levels you can also enter the weight of the group itself and choose how the within-group weights are rescaled. Two to four levels are supported, and four-level models with random slopes are limited to continuous outcomes. With few groups the group variances are very uncertain and are often estimated at zero.

📖 Reading the output

All tests use standard significance thresholds: *** p < .001, ** p < .01, * p < .05, ns p ≥ .05. Each result section ends with a plain-language interpretation paragraph suitable for direct use in a report.

Inferential tests ask is X related to Y? Effect Analysis asks something harder: did our programme cause this change? That is difficult because we can never see the same people both with and without the programme. The people who joined a programme are usually different from those who did not (more motivated, closer to the road, poorer), so simply comparing the two groups mixes the effect of the programme with the effect of those differences.

The four methods here are different ways of building a fair comparison group, which statisticians call the counterfactual: what would have happened to the participants if the programme had not existed. Each method rests on a different assumption, and the guide spells out what that assumption is, how to check it, and what to do when it is doubtful.

MethodThe idea in a sentenceUse when
Propensity Score MatchingFind non-participants who look just like participants, and compare those.One round of data, no random assignment, and good information on how participants and non-participants differ.
Difference-in-DifferencesCompare how much each group changed, not where each ended up.A programme and a comparison group, each measured before and after.
Regression Discontinuity DesignCompare people just above and just below an eligibility cut-off.Eligibility is decided by a score, age or income threshold.
Instrumental Variables (2SLS)Use something that nudges participation but has no effect on the outcome by any other route.Take-up is tied up with unmeasured factors and you have a credible instrument.
All four methods are under one Effect Analysis card on the Quantitative Analysis home screen, next to Inferential Analysis. If you also want to know how an effect works, through which middle step, see Mediation in Structural Equation Modelling.
No method can promise the right answer. Each estimate is only as good as its assumption. Read the diagnostics and sensitivity checks as seriously as the headline effect, and describe findings as consistent with an effect of that size, not as proof.

Propensity Score Matching

One round of dataBinary treatmentObserved confounders only
In plain words. Imagine each participant has a 'twin' among the people who did not take part: someone with the same age, schooling, household size and so on. Comparing outcomes for twins gives a fairer picture of the programme's effect than comparing everyone. A propensity score is each person's estimated chance of being in the programme, given their characteristics, and it lets AnalyZ match on all the characteristics at once instead of one at a time.
Use it when
  • Participation was not randomly assigned
  • You have only one round of data but good information on who joined and who did not
  • You can name the characteristics that drive both joining and the outcome
Your data needs
  • A binary treatment variable (1 = participated, 0 = did not)
  • A numeric or binary outcome
  • Covariates measured before the programme, for example age, sex, schooling. Covariates affected by the programme will bias the answer.

Propensity Score Matching: how to run it

1
Choose the Treatment variable (binary), Outcome variable and Covariates
The covariates are the characteristics used to estimate each person's propensity score.
2
Pick an Estimation technique and an Estimand
See the two tables below the results guide. A sensible start is nearest neighbor matching and ATT.
3
Click Run matching, then check balance before reading the effect
If matching has not made the groups similar, the effect estimate cannot be trusted.
4
Check the sensitivity analysis
Look at the E-value to judge how fragile the result is to something you did not measure.

Propensity Score Matching: reading the results

You seeIt means
Treatment effect estimate(s)The headline number: the average difference in the outcome between participants and their matched counterparts, with standard error, p-value and 95% CI.
Matching summaryHow many participants were matched, and how many controls were used or left out.
Covariate balance (SMD before and after)For each covariate, how different the two groups were before matching and after. An absolute standardised mean difference below 0.1 is good balance.
Matching qualityHow many covariates reached good balance, and the percentage of bias removed by matching.
Rubin's B and RA second check on the propensity scores themselves. B (standardised bias) should be under 25, R (variance ratio) between 0.5 and 2.
Propensity score model fitPseudo R² for the model that predicts participation. About 0.2 to 0.4 suggests your covariates separate participants from others reasonably well. This is not the quality of the effect estimate.
Sensitivity analysis (E-value)How strong an unmeasured factor would have to be, in its link to both participation and the outcome, to explain the whole effect away. Bigger is more reassuring.
Bias gridHow the estimate would shift under a range of hypothetical hidden confounding strengths.

Choosing an estimation technique

TechniqueHow it worksWhen to prefer it
Nearest neighbor matchingPairs each participant with the single most similar non-participant, as long as they are close enough (within the caliper). Participants with no close partner are dropped.You want a transparent, easy-to-explain method and have plenty of comparison people.
Kernel matchingBuilds each participant's counterfactual from all non-participants, giving more weight to the most similar. Nobody is dropped.You have few comparison people, or want to use all the data.
Inverse probability weighting (IPW)Re-weights the whole sample so that the two groups resemble each other. People who looked unlikely to join but did are given more weight.You want to use everyone and avoid explicit pairs.
Doubly robust (AIPW)Combines propensity weighting with an outcome regression. It gives a correct answer if either of the two models is right.You want protection against a mistake in one of the two models. A good default when sample size allows.

For nearest neighbor matching, the caliper is the largest allowed difference between a participant and their match. It is set as a share of the standard deviation of the propensity score on the logit scale, with 0.2 as the usual default. A tighter caliper gives closer matches but drops more people. For kernel matching you can set the bandwidth. IPW and doubly robust results use a bootstrap for their standard errors, so you set the number of Bootstrap replications (more means slower and steadier). For IPW you can also choose stabilised or unstabilised weights, and stabilised weights are less erratic.

Choosing an estimand

EstimandThe question it answersTypical choice when
ATTWhat was the effect on those who actually took part?Most evaluations. You want to know whether the programme worked for its participants.
ATCWhat would the effect have been on those who did not take part?You are considering extending the programme to non-participants.
ATEWhat is the effect across the whole population?The programme could be applied to anyone.
Worked example

The situation. A job-training programme enrolled 300 people. 700 others did not enrol. The outcome is monthly earnings six months later. Covariates: age, years of schooling, prior earnings and household size. Method: nearest neighbor matching, estimand ATT.

CovariateSMD beforeSMD after
Age0.350.04
Years of schooling0.410.06
Prior earnings0.520.05
Household size-0.12-0.03

What AnalyZ shows. ATT = +42 per month (95% CI 12 to 72, p-value 0.006). All four covariates have an absolute SMD below 0.1 after matching, so balance is good. Rubin's B = 12 and R = 1.1, both within the target. Propensity model pseudo R² = 0.12. E-value = 1.8.

How to read it. Before matching, participants were younger, better educated and had higher earnings before the programme than non-participants, so a raw comparison would have overstated the benefit. After matching the groups look alike. Participants earn about 42 more per month than comparable non-participants, and the interval stays above zero. The E-value of 1.8 says an unmeasured factor would need to be associated with both participation and earnings by a factor of about 1.8 to explain the whole effect away. That is moderate protection, not strong, because motivation, which you did not measure, could plausibly be that strong. Report the finding with that caveat.

Illustrative figures.

Propensity Score Matching: things to watch for

Propensity Score Matching: key terms

Propensity score
A person's estimated probability of being in the programme.
SMD
Standardised mean difference: a gap between groups in standard deviations.
Caliper
The largest allowed distance between a person and their match.
Counterfactual
What would have happened without the programme.

Difference-in-Differences

Two rounds of dataTreatment and comparison groups
In plain words. Compare how much each group changed between two points in time. Both groups may have risen anyway because prices went up or the season changed. The programme's effect is the extra change in the participants, beyond the change in the comparison group. The result is a 'difference of the two differences'.
Use it when
  • You have baseline and endline data on a programme group and a comparison group
  • You worry that the groups started at different levels (which DiD allows) but think they would have moved in parallel without the programme
Your data needs
  • A group variable (programme = 1, comparison = 0)
  • A round variable with two values (before and after)
  • A numeric or binary outcome
  • Optional covariates. Rounds can be in one long file or a second file you upload.

Difference-in-Differences: how to run it

1
Choose your Data structure
Already have both rounds (a round column in one dataset), or Upload a second round (baseline is the loaded dataset, endline a separate file, matched by ID).
2
Select the Round variable (or the ID variables and Round 2 file)
Several ID columns can be combined into one key when no single column is unique.
3
Choose the Group variable, Outcome variable and any Covariates
The group variable identifies programme versus comparison.
4
Click Run DiD and read the Group × Round line
That row is the effect.

Difference-in-Differences: reading the results

You seeIt means
Group, Round and Group × Round coefficients (robust SE)Group is the baseline gap between the groups. Round is the change over time in the comparison group. Group × Round is the programme effect: the extra change in the programme group.
Raw 2 × 2 differenceThe same calculation done directly from the four averages. It matches the regression exactly when there are no covariates, which makes it a useful cross-check.
Model summaryR², adjusted R² and the F statistic.
For a binary outcomeA logistic model is also fitted, and the effect is given as an average marginal effect, a change in probability, not the raw interaction term, which would not be the right DiD estimate.
Parallel trends. DiD assumes the two groups would have moved by the same amount without the programme. With only two time points this cannot be tested. Check it with whatever you know: were the groups on similar trajectories before? Did anything else change for one group only? The subsample filter does not apply when you upload a second-round file.
Worked example

The situation. A savings-group programme ran in some villages. Household income was measured before and after, in 150 programme and 150 comparison households.

GroupBeforeAfterChange
Programme100130+30
Comparison98112+14
Difference-in-differences+16

What AnalyZ shows. Group × Round = 16.0 (robust SE 5.2, p-value 0.002, 95% CI 5.8 to 26.2). The raw 2 × 2 estimate is also 16.

How to read it. Income rose in both groups, by 14 in the comparison villages simply from the passage of time. Programme households gained 30, which is 16 more. The extra 16 is the estimated programme effect. A simple before-and-after look at the programme group alone would have claimed 30, nearly double the credible figure. A simple end-of-programme comparison would have claimed 18 (130 vs 112), which wrongly ignores the 2-point head start the programme group had.

Illustrative figures.

Difference-in-Differences: things to watch for

Difference-in-Differences: key terms

Parallel trends
The assumption that both groups would have followed the same path without the programme.
Interaction
Here, the extra change belonging to the programme group in the later round.
Baseline and endline
The first and last measurements.

Regression Discontinuity Design

Eligibility cut-offSharp or fuzzy
In plain words. When a programme is offered only to those who score above (or below) a cut-off, people just either side of the line are almost identical. A student who scored 59 is much like one who scored 61. Yet one got a scholarship and the other did not. Any sudden jump in the outcome right at the cut-off is then very likely caused by the programme.
Use it when
  • Eligibility is decided by a threshold on a score, index, income level or age
  • You have data on people near the threshold
Your data needs
  • A running variable: the continuous score that defines the cut-off
  • A numeric or binary outcome
  • Enough people close to the cut-off
  • For a fuzzy design, a variable showing who actually received the programme

Regression Discontinuity Design: how to run it

1
Choose Sharp or Fuzzy
Sharp: everyone on one side was treated, with no exceptions. Fuzzy: some people above the cut-off did not take up the programme (or some below did). Fuzzy needs the Actual treatment received variable and estimates a local average effect (LATE) using 2SLS.
2
Set the running variable, outcome and Cutoff value
The cutoff defaults to the median of the running variable. Change it to your real threshold.
3
Choose the Bandwidth
Only people within this window of the cut-off are compared. Manual lets you set the width in the running variable's own units. Automatic chooses it from the data by cross-validation.
4
Optionally add covariates to check
Numeric background variables that should not jump at the cut-off.
5
Click Run RDD, then read every diagnostic before the headline estimate
A convincing jump needs a credible design.

Regression Discontinuity Design: reading the results

You seeIt means
Treatment effect at the cutoffThe size of the jump in the outcome at the threshold: the effect for people right at the line.
Manipulation check (density at the cutoff)Tests whether unusually many people sit just above or just below the cut-off, a sign that scores were manipulated or people sorted themselves. A significant result is a warning.
Sorting check (counts within the bandwidth)Compares how many observations are just below and just above.
Covariate balance at the cutoffBackground variables should not jump at the cut-off. If they do, something besides the programme changes there too.
Bandwidth sensitivityThe estimate re-run with windows from half to double the chosen width. A stable estimate is reassuring. Big swings mean the result is fragile.
Placebo cutoff testThe same test at fake cut-offs where nothing should happen. A significant placebo is a warning.
First stage (fuzzy only)How strongly the cut-off predicts who actually got the programme. If the jump is small, the estimate is unstable.
Model fitR², adjusted R² and F for the local regression within the bandwidth.
AnalyZ fits a straight line on each side of the cut-off, within the bandwidth. This is the specification recommended by current methodology (Gelman and Imbens, 2019), because high-order curves give noisy estimates that depend on the chosen shape. The outcome must be numeric or binary. A nominal outcome with three or more categories is not supported.
Worked example

The situation. A scholarship goes to students with an entrance score of 60 or more. The outcome is a test score two years later. A sharp design, automatic bandwidth (8 points either side of 60).

What AnalyZ shows. Effect at the cutoff = +6.2 points (SE 2.1, p-value 0.003). Manipulation test: not significant (p-value 0.47). Covariate balance for family income and age: no jump (p-values 0.61 and 0.38). Bandwidth sensitivity: estimates from 4.9 to 7.0 across half to double the bandwidth. Placebo cutoffs at 50 and 70: not significant.

How to read it. Students who just scraped in did about 6 points better two years later than those who just missed out. The checks all support this. The scores do not bunch around 60, so nobody seems to have gained access by manipulation. Family income and age do not jump at the line, so the two sides really are alike. The estimate holds from 4.9 to 7.0 when the window is changed, and fake cut-offs produce nothing. The estimate applies to students near the cut-off, not to high or low scorers.

Illustrative figures.

Regression Discontinuity Design: things to watch for

Regression Discontinuity Design: key terms

Running variable
The score that decides eligibility.
Cutoff
The threshold value.
Bandwidth
How close to the cutoff someone must be to be used.
LATE
Local average treatment effect: the effect for those whose treatment was changed by crossing the cutoff.

Instrumental Variables (2SLS)

Unobserved confoundingNeeds an instrument
In plain words. Sometimes the people who take up a programme differ from others in ways you cannot measure, so matching cannot fix the comparison. An instrument is a variable that nudges people into the programme but affects the outcome by no other route, like a random invitation letter or distance to the programme site. The method looks only at the part of participation that the nudge explains, which is free of hidden motives, and traces what happens to the outcome. It is run in two steps, hence two-stage least squares (2SLS).
Use it when
  • Participation is tied up with unmeasured factors such as motivation, ability or reverse causation
  • You have a credible instrument you can defend on substantive grounds
Your data needs
  • A numeric or binary outcome
  • A numeric or binary endogenous treatment variable
  • At least one instrument
  • Optional exogenous covariates

Instrumental Variables: how to run it

1
Choose the Outcome variable and the Endogenous treatment variable
Binary variables are handled as a linear probability model.
2
Choose the Instrument(s)
It should affect the treatment but not have a direct effect on the outcome. More than one instrument unlocks an extra check but each still needs its own justification.
3
Add Exogenous covariates (optional)
Background variables that are not themselves affected by the programme.
4
Click Run 2SLS and check the first-stage F-statistic before anything else
If the instrument is weak, nothing downstream is reliable.

Instrumental Variables: reading the results

You seeIt means
First-stage strength (F statistic)How strongly the instrument predicts the treatment. Above 10 is the usual 'not weak' threshold. Below that the 2SLS estimate becomes unreliable.
2SLS estimateThe headline effect of the treatment on the outcome, using only the variation the instrument creates.
OLS versus 2SLSThe naive estimate beside the corrected one. A big gap shows how much the hidden factors were distorting the simple comparison.
Reduced formThe outcome regressed directly on the instrument, which is a building block of the final estimate.
Endogeneity test (Durbin-Wu-Hausman)Whether the treatment really is tied up with hidden factors. A significant result supports using 2SLS rather than ordinary regression.
Overidentification test (Sargan)Only shown with more than one instrument. Tests whether the instruments agree with each other. A significant result suggests at least one may be invalid.
Model fitFirst-stage R² and adjusted R², with a 2SLS R² for reference only, because it is not a reliable quality measure for 2SLS.
The key assumption cannot be tested. No statistic can prove an instrument affects the outcome only through the treatment. The Sargan test only checks whether several instruments agree with each other. Whether an invitation letter, distance or a lottery is a valid instrument is a judgment about your programme. Argue it in your write-up.
Worked example

The situation. A financial-skills workshop was open to all, but only some chose to attend. The programme also mailed invitation letters to a randomly chosen half of households. Instrument: received an invitation. Outcome: monthly income. Treatment: attended the workshop.

What AnalyZ shows. First-stage F = 38. OLS effect = +85 per month. 2SLS effect = +52 per month (SE 18, p-value 0.004). Endogeneity test p-value 0.03.

How to read it. The invitation strongly predicted attendance (F = 38, well above 10), so it is not a weak instrument. The naive OLS estimate of +85 is much larger than 2SLS (+52), and the endogeneity test confirms that attendance was tied up with hidden factors. Plausibly, motivated people both attend and earn more. Correcting for this brings the effect down to +52. This applies to people whose attendance was influenced by the invitation.

Illustrative figures.

Instrumental Variables: things to watch for

Instrumental Variables: key terms

Instrument
A variable that affects take-up but not the outcome directly.
Endogenous
Tied up with hidden factors that also affect the outcome.
Exclusion restriction
The assumption that the instrument affects the outcome only through the treatment.
First stage
The step that predicts treatment from the instrument.

Survey weights

When a survey weight is set in Data View, all four methods use it by default. Untick Apply survey weight beside Run to compare with the unweighted result. Estimates are weighted, and uncertainty comes from the survey design (weights, strata and clusters) in place of the usual robust or bootstrap errors.

MethodWhat changes
Propensity Score MatchingThe propensity model is fitted with the survey weights, and matching, the balance checks (standardised differences and Rubin's B and R) and the effect estimates carry the weights. Nearest-neighbour matching uses the design variance of the weighted mean of matched differences. Kernel matching, inverse probability weighting and the doubly robust estimate use a bootstrap that resamples the survey's clusters within strata and re-estimates the propensity model each time. Weighted N for the treated and control groups appears in the model table.
Difference-in-DifferencesThe effect comes from weighted regression with design-based standard errors, and the cell means in the 2x2 table are weighted. For a binary outcome the weighted logit and the marginal DiD are shown, with the marginal effect's uncertainty from a bootstrap of the survey's clusters. The weight cannot be used when the second round is uploaded as a separate file, so put both rounds in one dataset.
Regression DiscontinuityA weighted local linear regression within the bandwidth, with design-based standard errors. The cross-validated bandwidth, density check, covariate balance, bandwidth sensitivity and placebo cutoffs all use the weights.
Instrumental Variables (2SLS)The first stage, the 2SLS and naive OLS estimates and the reduced form are weighted, with design-based standard errors. The first-stage F statistic is a design-based Wald F, and the endogeneity test uses the same errors. The over-identification (Sargan) test is not shown with a survey weight, because it has no accepted design-based version.

📖 Reading the output

All four methods report standard significance thresholds, a plain-language interpretation, and a model fit table appropriate to that method. For Propensity Score Matching, always read the balance diagnostics and sensitivity analysis alongside the point estimate. For Regression Discontinuity Design, treat the headline estimate as provisional until the manipulation test, covariate balance, and bandwidth sensitivity checks all come back clean. For Instrumental Variables, check the first-stage F-statistic before reading anything else, a weak instrument undermines every downstream number.

Many things we care about, such as trust, wellbeing or food security, cannot be measured with one question. We ask several questions and treat them as signs of an underlying idea, called a factor or construct. Structural Equation Modelling (SEM) does two jobs together: it checks that your questions really measure the idea you intended (the measurement part), and it tests how those ideas affect each other (the structural part). Because it allows for measurement error, it gives cleaner effect estimates than averaging items and running a regression.

AnalysisUse it whenStart here if you are new
Confirmatory Factor AnalysisYou want to check that a set of questions measures the factors you expect.Yes. Always check your scales here first.
Structural Equation ModelYou want to test paths between factors (or between factors and measured variables).After your CFA is sound.
Compare GroupsYou want to know whether a scale means the same thing in every group, or whether effects differ by group.Before comparing groups on a scale.
Path AnalysisYou have measured variables only and want direct and indirect effects.When you have no multi-item scales.
Mediation AnalysisYou want to test whether X affects Y through one mediator M.Simple, all-numeric chains.
Counterfactual MediationA mediator or outcome is binary or ordered, or you need covariates, several mediators or group results.Anything beyond the simple chain.
Beta. This module is new. Results follow established methods, but check important findings against another package before you publish, and tell us if something looks wrong.

Things that apply to every analysis

TopicWhat to know
Numeric variablesAll variables must be numeric. Likert items stored as numbers work well.
Variable typesChoose All continuous to treat items as numeric scores, or Mix of binary, ordinal and continuous to treat each by its number of different values: 2 is binary, 3 to 10 is ordinal, more than 10 is continuous. A table under the model shows how each variable will be treated.
Categorical predictorsChoose Categorical to code the variable into one 0/1 variable per category against a reference category you pick, or Continuous to use its values as numbers.
Survey weightsIf you set a survey weight or design in Data View it is used and the weighted N is reported. Bootstrap intervals then resample whole clusters.
Subsample and missing dataThe subsample filter applies. Only complete cases on the variables in the model are used, and the number used is shown with the results.
SharingAdd to Canvas lets you pick which chart to send. Add to Sense Making sends the tables and interpretation. Word and Excel exports include all tables and charts.
Sample sizeAim for at least 10 cases per estimated parameter, never fewer than 100 cases, and at least 200 for structural equation models.

Reading model fit

A model is a simplified story about how variables relate. Fit measures how closely that story reproduces the relationships actually seen in your data. A poor fit means the story leaves out something important.

MeasureGood fitIn plain words
CFI and TLI0.95 or higherHow much better your model is than a model that assumes nothing is related. Closer to 1 is better.
RMSEA0.06 or lowerThe average size of the mismatch between model and data, allowing for model simplicity. Smaller is better.
SRMR0.08 or lowerThe average gap between the correlations in your data and those the model implies. Smaller is better.
The chi-square test of fit almost always rejects a model in a large sample, even when the model is close. Judge fit mainly by CFI, TLI, RMSEA and SRMR, and look at all of them together rather than one.

Confirmatory Factor Analysis

Check a scaleTheory-drivenBeta
In plain words. You decide in advance which questions belong to which factor, and CFA checks how well your data agree. If five questions are meant to measure wellbeing, CFA tests whether they move together as one idea and whether each question is a good signal of it. Think of it as a quality check on your questionnaire.
Use it when
  • You use a multi-question scale and want proof it works
  • You want reliability and validity evidence for a report
  • You need to confirm two factors are really distinct
Your data needs
  • Numeric items, such as Likert answers
  • At least two items per factor (three for a single factor)
  • Each item belongs to one factor
  • At least 100 cases, and 10 per estimated parameter

Confirmatory Factor Analysis: how to run it

1
Define your factors
Name each factor and choose its items. Add more factors as needed. Decide the structure from theory or questionnaire design, not from the data.
2
Choose the variable types
All continuous, or a mix of binary, ordinal and continuous for items such as Likert responses.
3
Click Run Confirmatory Factor Analysis and read in order
Fit first, then loadings, reliability and discriminant validity, then suggested changes.

Confirmatory Factor Analysis: reading the results

You seeIt means
Fit summary (rated good, acceptable or poor)Whether the structure you specified matches the data. See the fit table above.
Standardised loadingsHow strongly each item reflects its factor, like a correlation. 0.50 or higher is reasonable. Below 0.40 the item is a weak signal.
Composite reliabilityHow consistently the items measure the factor. 0.70 or higher is acceptable.
Average variance extracted (AVE)How much of the items' variation is captured by the factor rather than noise. 0.50 or higher is good.
Discriminant validity (HTMT)Whether two factors are really different. A ratio above 0.85 means respondents are probably not telling them apart.
Suggested changesPlaces where letting two items' errors relate would improve fit. Add them only if the question wording makes sense of it, and report every change.
Worked example

The situation. A 6-question scale: three questions on trust in health workers and three on trust in local government, answered by 420 people on a 1 to 5 scale. Two factors, three items each.

ItemFactorStandardised loading
Trust1Health workers0.78
Trust2Health workers0.81
Trust3Health workers0.69
Trust4Government0.74
Trust5Government0.80
Trust6Government0.43

What AnalyZ shows. CFI 0.97, TLI 0.95, RMSEA 0.05, SRMR 0.04. Composite reliability 0.83 (health) and 0.76 (government). AVE 0.58 (health) and 0.50 (government). HTMT 0.62.

How to read it. Fit is good on all four measures. Five items load at 0.69 or more. Trust6 loads at 0.43, below the 0.50 guide, so it is a weak signal of government trust, though the factor is still acceptably reliable and its AVE just reaches 0.50. HTMT of 0.62 is well under 0.85, so the two kinds of trust are distinguishable. Next step: look at the wording of Trust6 and consider dropping it. Re-run to confirm fit and reliability hold.

Illustrative figures.

Confirmatory Factor Analysis: things to watch for

Confirmatory Factor Analysis: key terms

Factor
An underlying idea measured through several questions.
Loading
How strongly an item reflects its factor.
AVE
Average variance extracted: the share of item variation the factor explains.
HTMT
A ratio showing how distinct two factors are.

Structural Equation Model

Factors and pathsNeeds 200+ casesBeta
In plain words. Once your scales check out, the structural part asks how the ideas affect each other: does trust lead to service use, and does service use lead to wellbeing? It draws these as arrows (paths) and estimates each arrow's strength, while allowing for the fact that each factor is measured imperfectly. Ordinary regression on averaged scores ignores that imperfection and tends to understate effects.
Use it when
  • You have a theory with several linked constructs
  • You want effects that are not shrunk by measurement error
Your data needs
  • Everything CFA needs
  • At least 200 cases
  • A clear theory about which factor affects which

Structural Equation Model: how to run it

1
Define the factors
As in CFA. If unsure the measurement holds, run CFA first.
2
Define the paths
Add one equation for each factor or variable that others predict. The outcome and predictors can be factors or measured variables. Leave out paths you expect to be zero. Including every path makes the structural part fit perfectly by construction, so fit tells you nothing.
3
Click Run Structural Equation Model and read in order
Fit, then loadings and reliability, then structural paths, then the route-by-route effects.
4
Optionally compare models
Save a model, change it and run again to see whether a simpler model fits worse.

Structural Equation Model: reading the results

You seeIt means
Structural pathsEach arrow's estimate with p-value. The standardised version tells you relative strength: about 0.1 small, 0.3 moderate, 0.5 large.
R² for each outcomeHow much of that factor the model explains.
Effects route by routeEach way one factor reaches another: direct, through each middle factor, and total.
Residual diagnostics and suggested changesRelationships the model leaves unexplained. Large ones suggest a missing path.
DiagramDrag boxes to tidy it. Download any chart as PNG or SVG.
Worked example

The situation. Using the two trust factors from the CFA example, test whether trust in health workers predicts a service use factor (4 items), which in turn predicts wellbeing (5 items). N = 420.

PathStandardised effectp-value
Health-worker trust → Service use0.46< 0.001
Service use → Wellbeing0.34< 0.001
Health-worker trust → Wellbeing (direct)0.120.09

What AnalyZ shows. CFI 0.95, RMSEA 0.05, SRMR 0.05. R² = 0.21 for service use and 0.18 for wellbeing.

How to read it. Fit is acceptable to good. Trust has a fairly strong effect on service use (0.46), and service use has a moderate effect on wellbeing (0.34). The direct path from trust to wellbeing is small and not significant, which suggests that trust works through service use rather than directly. The model explains about a fifth of each outcome, so much remains unexplained. To formally test the indirect route, use Mediation.

Illustrative figures.

Structural Equation Model: things to watch for

Structural Equation Model: key terms

Path
An arrow from one variable or factor to another.
Structural model
The part of the model that links factors.
Direct and indirect effect
The part that goes straight, and the part that passes through a middle variable.

Compare Groups

Measurement invariance2 to 8 groupsBeta
In plain words. Before comparing, say, women and men on a scale, you must be sure the scale means the same thing to both. If the questions are understood differently, any gap in scores is not a real gap. Measurement invariance tests this step by step. The same tool can also test whether the effects in a model differ between groups.
Use it when
  • You want to compare scale scores across countries, sexes or programme groups
  • You want to know whether a path is stronger in one group than another
Your data needs
  • A group variable with 2 to 8 groups
  • At least 30 cases per group
  • Items that already pass CFA overall

Compare Groups: how to run it

1
Choose what to test
Measurement invariance (does the scale work the same way?) or Equal paths (do effects differ?).
2
Choose the group variable and the variables or model
For items answered as Likert or yes/no, choose Ordered categories. Loadings and thresholds are then tested together.
3
Click Run Compare Groups and read the levels from top to bottom
Stop at the first level that is not supported.

Compare Groups: reading the results

You seeIt means
Configural invarianceSame structure in every group. If supported, groups understand the scale in the same overall way.
Metric invarianceLoadings are equal. If supported, you can compare correlations and paths across groups.
Scalar invarianceLoadings and item intercepts are equal. If supported, you can also compare group averages on the factor.
Change in CFI, RMSEA, SRMRA level is supported when fit barely changes from the one before. This matters more than the chi-square difference, which is nearly always significant in big samples.
Partial invariance tableShows which loadings or intercepts differ most. You may free a few and re-run.
Worked example

The situation. The six-item trust scale was compared between women (n = 230) and men (n = 190).

LevelCFIChange in CFIVerdict
Configural0.970Supported
Metric0.968-0.002Supported
Scalar0.951-0.017Not supported

How to read it. The scale has the same structure and the same loadings in both sexes, so you can compare how trust relates to other variables. But the scalar level fails, because CFI dropped by 0.017 (a drop above about 0.01 is the usual warning). That means that you cannot yet compare average trust between women and men. The partial invariance table points to Trust6 as the main culprit. Freeing its intercept and re-running may restore scalar invariance partially, which is usually enough. If many items needed freeing, the scale would not be measuring the same thing in both groups and no comparison would be safe.

Illustrative figures.

Compare Groups: things to watch for

Compare Groups: key terms

Measurement invariance
The scale measures the same thing in every group.
Intercept
An item's baseline level when the factor is zero.

Path Analysis

Measured variablesDirect and indirect effectsBeta
In plain words. A set of regressions run together, with arrows between your measured variables. A variable can be an outcome in one equation and a predictor in the next, which creates indirect effects: schooling affects income partly through job type.
Use it when
  • You have several measured outcomes linked in a chain
  • You want direct, indirect and total effects
  • Some outcomes are binary or ordered
Your data needs
  • Numeric variables, treated as measured without error
  • A clear theory of which variable predicts which

Path Analysis: how to run it

1
Choose the variable types
With All continuous, a two-value outcome is modelled with probit and can only be the last step in a path. With Mix of binary, ordinal and continuous, such variables can sit anywhere in the chain.
2
Add one equation per outcome
Give each outcome and the variables that predict it. Leave out paths you expect to be zero.
3
Click Run Path Analysis and read the diagram and the tables
The diagram shows each path with its standardised effect.

Path Analysis: reading the results

You seeIt means
EstimatesEach path's strength.
Variance explainedR² for each outcome.
Direct, indirect and total effectsAn indirect effect is significant when its 95% interval does not include zero.
Worked example

The situation. Do years of schooling affect income directly, or through the type of job a person gets? Variables: schooling, job skill level (1 to 5), income. N = 300.

Effect of schooling on incomeStandardised95% CI
Direct0.210.09 to 0.33
Indirect (via job skill level)0.280.18 to 0.38
Total0.490.39 to 0.59

How to read it. Schooling raises income by two routes. About 43% of the total (0.21 of 0.49) is direct. The rest, 0.28, runs through getting a higher-skilled job. Both intervals exclude zero, so both routes matter. Because schooling has a sizeable effect on income even after taking job skill into account, job type alone does not explain the benefit.

Illustrative figures.

Path Analysis: things to watch for

Path Analysis: key terms

Indirect effect
An effect that passes through a middle variable.
Total effect
Direct plus indirect.

Mediation Analysis

One mediatorAll numericBootstrap
In plain words. Asks how or why something works. If a training programme (X) raises income (Y), is it because it raises skills (M)? Mediation splits the total effect into the part that travels through the mediator (indirect) and the part that does not (direct).
Use it when
  • You want to open the black box of how a programme works
  • You have one mediator and everything is numeric
Your data needs
  • X, M and Y, all numeric and measured on the same cases
  • A sensible time or causal order: X before M before Y

Mediation Analysis: how to run it

1
Choose X, the Mediator and Y
In causal order.
2
Set the number of resamples and a seed
Use 1,000 to 5,000 resamples. A fixed seed makes the result reproducible.
3
Click Run Mediation Analysis and read the indirect effect first
Judge it by its bootstrap interval. The Sobel test is also shown but the bootstrap is more reliable.

Mediation Analysis: reading the results

You seeIt means
Total, direct and indirect effectsTotal = direct + indirect.
Bootstrap 95% CI for the indirect effectSignificant if the interval does not include zero.
Sobel testAn older test of the same thing. Less reliable than the bootstrap.
Proportion mediatedThe share of the total effect that goes through M. Unstable when the total effect is close to zero, so report it only when the total effect is clearly non-zero.
Type of mediationFull: indirect significant, direct not. Partial: both significant. Suppression: opposite signs, so interpret with care. None: indirect not significant.
Worked example

The situation. Does a training programme (hours attended) raise income through skill score? N = 250.

EffectEstimate95% bootstrap CI
Total4.83.1 to 6.5
Direct1.90.2 to 3.6
Indirect (via skills)2.91.8 to 4.2

What AnalyZ shows. Sobel p-value < 0.001. Proportion mediated = 60%. Partial mediation.

How to read it. Each extra training hour is linked to 4.8 more in income. About 2.9 of that (60%) travels through improved skills, and 1.9 does not. Both parts are significant, so this is partial mediation: skills explain a large part of the benefit but not all of it. Something else the training does, such as networking, may account for the rest.

Illustrative figures.

Mediation Analysis: things to watch for

Mediation Analysis: key terms

Mediator
The middle step on the route from X to Y.
Bootstrap
Re-running the analysis on many resampled versions of your data to see how much the answer varies.

Counterfactual Mediation

Binary or ordered variablesCovariatesSensitivity checksBeta
In plain words. The same question as Mediation, but built to handle real-world data. It works with binary and ordered variables, covariates, several mediators, a categorical exposure and group-by-group results. It simulates what the outcome would be under different combinations of exposure and mediator values, and reports the direct and indirect effects.
Use it when
  • The mediator or outcome is binary or ordered
  • You need to control for covariates
  • You have 2 or 3 mediators
  • You want effects for each group or for each category of exposure
Your data needs
  • An exposure, one or more mediators, an outcome, and optional covariates
  • Include covariates that could affect both the mediator and the outcome

Counterfactual Mediation: how to run it

1
Set up the causal chain
Choose the exposure, mediator(s), outcome and covariates. Mark ordered variables as ordered categories.
2
Choose the comparison
For a binary exposure it compares the higher with the lower value. For a continuous one it compares a 1 standard deviation increase (or 1 unit) from the average. Then choose Everyone together, A categorical exposure (3 to 10 categories, with a reference you choose), or Each group separately (2 to 6 groups).
3
Click Run Counterfactual Mediation and read the effects
An effect is significant when its 95% bootstrap interval excludes zero. Use a fixed seed.
4
Open the sensitivity panel
See how strong hidden confounding would need to be to change your conclusion.

Counterfactual Mediation: reading the results

You seeIt means
Total, direct and indirect effectsFor a binary outcome the effects are in percentage points. An ordered outcome is shown as the probability of reaching a category or higher, or the expected score.
Several mediatorsParallel mediators do not affect each other. A serial model has exactly two, the first affecting the second, and splits the route through the first, the second and both.
Categorical exposureEach other category is compared with your reference, with its own total, direct and indirect effect. Comparisons are not adjusted for how many you make.
Each group separatelyThe whole model is fitted within each group. Each group is compared with a reference group, with intervals from paired resamples.
E-valuesThe smallest strength of link a hidden factor would need with both sides of the relationship to explain the effect away. Bigger is more reassuring. For direct and indirect effects they are a guide only.
Hidden-correlation curveFor one continuous mediator and outcome, shows the correlation between hidden factors at which the indirect effect would fall to zero.
Worked example

The situation. Does a cash transfer (binary: received or not) raise the chance a child is enrolled in school (binary) through household food security (binary mediator)? Covariates: household size and head's schooling. N = 800.

Effect on school enrolmentPercentage points95% bootstrap CI
Total+9.0+4.1 to +13.8
Direct+5.4+0.3 to +10.4
Indirect (via food security)+3.6+1.5 to +5.9

What AnalyZ shows. E-value for the total effect = 1.7.

How to read it. Receiving the transfer raises enrolment by about 9 percentage points. Roughly 3.6 of those points (40%) work through improved food security, while 5.4 do not. An E-value of 1.7 suggests only a moderately strong hidden factor, linked to both receiving the transfer and enrolment, would explain the effect away, so keep this in mind in your write-up and do not claim certainty.

Illustrative figures.

Counterfactual Mediation: things to watch for

Counterfactual Mediation: key terms

Counterfactual
What would have happened under a different exposure.
E-value
How strong an unmeasured factor must be to explain an effect away.
Serial mediation
Two mediators in a chain, the first affecting the second.

Open Modules in the top bar, then Data Visualization. The screen has a Chart type panel on the left, the settings for the chart you choose in the middle, and your chart below the Generate button. You need a dataset loaded first.

Build your first chart, step by step

1
Open a group in the Chart type panel
The six groups are Compare groups, Trends, Proportions, Relationships, Distributions and Specialised. Click a group name to open it, then click the chart you want, for example Bar. A short hint under the group says what it is for.
2
Choose your variables
The settings in the middle depend on the chart. For a Bar chart choose the variable to count. Most charts also offer an optional by or colour variable that splits the chart into groups. Some settings, such as Stack type, Summarise as (Mean or Median), Sort by and Max rows shown, appear only for charts that need them.
3
Click Generate
The chart appears below. Generate stays off until the required variables are chosen. If a survey weight is set, an Apply survey weight switch sits beside the button.
4
Explore it
Hover for exact values. Drag to zoom, double-click to reset. The toolbar at the top right of the chart has a camera icon that saves a PNG picture.
5
Change the look
Open 🎨 Edit chart under the chart. Change the Title, X-axis label, Y-axis label, Font size, Background, Text colour, Legend position and the axis limits (X min, X max, Y min, Y max), then click Apply. ↺ Reset goes back to the defaults.
6
Look at part of your data
Open Filter data, pick a Column and a Value, add more rows with + Add condition, then click Apply & Regenerate. ✕ Clear filter removes it.
7
Write what the chart shows
In Interpretation, type one or two sentences, or use the AI option. Click Save to keep it with the chart.
If you change the chart type or variables after generating, a warning asks whether to replace the existing chart. Click Yes, replace it to continue or Cancel to keep the old one.

Buttons above the chart

ButtonWhat it does
HTMLDownloads this chart as an interactive web page you can open in any browser.
Add to Sense MakingSaves the chart and a summary of its data into a Sense Making project, where the AI can help you interpret it. Choose an existing project or create one, and give the item a title.
Add to CanvasAdds the chart as a slide in a Data Canvas story. Write the slide headline and choose a canvas, or create one.
Add to dashboardPuts the chart on your dashboard so it can be shown with others. See below.

Available chart types

The groups below are the ones shown in the Chart type panel.

Compare groups

How do categories differ from each other?

ChartBest forVariables needed
BarFrequency or value of a category1 category (+ optional split)
Stacked BarPart-of-whole comparison across groups1 category + 1 group variable
ScatterRelationship between two numeric variables2 numeric
TreemapRelative size of groups (tile area is the number of rows per category)1 category (+ optional second grouping)
BubbleThree-variable relationships (size encodes a third variable)3 numeric
QuadrantTwo-axis positioning with centre reference lines2 numeric + 1 label
Range chartMin–max or CI ranges across categories1 label + 2 numeric (lo/hi)

Trends

How does a value change over time or sequence?

ChartBest forVariables needed
LineChange over time or ordered categories1 X column + 1 numeric
AreaCumulative trends over time1 X column + 1 numeric
StreamgraphStacked proportional flow over time by categoryX + numeric + 1 category
SparklineMinimal trend line with min, max and latest value marked; one per group if you split1 numeric (+ optional order variable, + optional split)

Proportions

How does each part contribute to the whole?

ChartBest forVariables needed
PieProportions of a whole1 category variable
DonutSame as pie, modern hollow style1 category variable
SunburstHierarchical part-of-whole with nested rings1 label + optional parent column

Relationships

How are variables related or correlated?

ChartBest forVariables needed
HeatmapCorrelation matrix across multiple variables2+ numeric
ContourDensity surface over two numeric variables2 numeric
RadarMultivariate comparison on radial axes1 category + 1 numeric
Radial BarCircular bar chart for part-of-whole comparisons1 category + 1 numeric

Distributions

How is a variable spread across its range?

ChartBest forVariables needed
HistogramDistribution of a single numeric variable1 numeric
BoxplotSpread and outliers across groups1 numeric (+ optional group)
ViolinFull distribution shape across groups1 numeric (+ optional group)
Density plotSmooth probability density curve2 numeric
Radial histogramCircular frequency distribution (e.g. directional data)1 numeric

Specialised

Niche chart types for specific use cases.

ChartBest forVariables needed
Circular gaugeSingle KPI against a target value1 numeric
Forest plotEffect sizes with confidence intervals (meta-analysis)Label + effect + CI lo/hi columns
SankeyFlow between two sets of categories2 category columns
TableA formatted table of selected columns, with optional sortingAny columns, with optional sort column and maximum rows shown (50 by default)
Multiple ResponseFrequency of each option in a "select all that apply" question, optionally by groupEither one column with answers separated by a delimiter (such as ;) or one Yes/No (1/0) column per option

Survey weights

When a survey weight is set in Data View, charts use it by default. Untick Apply survey weight beside Generate to see the unweighted chart. The weight is saved with each chart, so charts you add to a dashboard, filter or share keep the setting they were built with.

ChartsWhat changes
Bar, Stacked Bar, Pie, Donut, Treemap, Sunburst, Sankey, Multiple Response, Radial histogramEach respondent counts with their weight, so counts are weighted counts (estimates for the population) and percentages are weighted. Axes and tooltips say "Weighted count".
Histogram, Density plot, ContourEach value adds its weight to its bin, so the heights show weighted frequency.
Line, Area, Streamgraph, Radar, Radial Bar, Circular gauge, and the mean overlay on a Bar chartMeans, medians and totals are weighted. Axis titles read, for example, "Weighted mean of income".
Boxplot, ViolinDrawn from the weighted distribution, so the quartiles, median and mean are weighted. The violin smoothing uses the effective sample size implied by the weights.
Heatmap, Scatter trend lineCorrelations are weighted, and Spearman uses weighted ranks. The trend line is a weighted least squares line.
Scatter points, Bubble, Quadrant, Range chart, Forest plot, Table, Sparkline lineNot changed, because they show one point or row per respondent and nothing is counted or averaged. The Sparkline reference line is the weighted mean.

Build and share a dashboard

1
Build a chart and click Add to dashboard
A confirmation says Chart added to dashboard! The same chart cannot be added twice. A tab for it appears above the builder.
2
Build more charts
Click + New chart in the tab row, make the next chart, and add it. Click a chart's tab to reopen it, or the × on the tab to remove it from the builder and the dashboard.
3
Click View dashboard
It sits in the Dashboard bar with My Dashboards and Save & Share. The charts appear in a grid.
4
Arrange the layout
On each chart you can drag the handle to reorder, choose Half width or Full width, and Move to page. Add page creates another page, and Delete page removes one. Add more charts takes you back to the builder.
5
Cross-filter
Click any bar, slice or point. The other charts update to show only that group. Clear cross-filter resets.
6
Interpret the charts
✦ Interpret all writes a short plain-language reading for each chart. The counter beside it, for example 1/1 interpretations left, shows your remaining monthly allowance.
7
Download
Use Save as HTML in the dashboard to get one interactive file you can email or open offline.

Save & Share

1
Click Save & Share
It is in the Dashboard bar in the builder.
2
Fill in the details
Enter the Dashboard name, an optional One-line summary (up to 160 characters) and an optional Description / context that appears at the top of the shared page.
3
Choose who can see it
The switch reads Private, only you can view or Public, anyone with the link can view. Public sharing needs a plan that includes it. Otherwise you will see You can save this dashboard, but public sharing needs a paid plan. with an Upgrade button.
4
Pick filter variables (optional)
Under Filterable variables, choose the columns visitors may filter by. Click Select filter variables.
5
Save, then copy the link
Click the save button, then Copy to copy the link. Saving again over an existing dashboard asks Yes, update or Save as new.

My Dashboards lists dashboards you saved. Click Load beside one to reopen it for editing. If the dataset is still loaded, every chart can be rebuilt. Without the dataset, the charts show their saved pictures only.

Which chart should I use?

Your questionGood chartWhy
How do categories compare (for example % by district)?Bar chartLength is the easiest thing to judge by eye. Sort the bars from largest to smallest.
How does something change over time?Line chartThe slope shows direction and speed. Use one line per group.
What share of the whole does each part make up?Stacked bar, or a pie or donut with few partsKeep to 5 parts or fewer, or the slices become hard to compare.
Are two numbers related?Scatter plotEach dot is one person. A tilt shows a relationship, and a cloud shows none.
What does the spread of a number look like?Histogram or box plotShows the shape, skew and outliers that an average hides.
How do groups differ on a numeric variable?Box plot by group, or grouped bars of the meanBox plots show spread as well as the typical value.
Reading a box plot. The box covers the middle half of the people, the line inside is the median, and the whiskers show the usual range. Dots beyond the whiskers are unusual values worth checking. Reading a histogram. Taller bars mean more people in that range. A long tail to the right means a few very high values.

A scale is a set of questions that together measure one idea, such as stress, trust or satisfaction. Adding the answers into one score is more dependable than relying on a single question, but only if the questions really do belong together. This module checks that. Choose a card, open the analysis, and use Back to return to the card selection. Results can be exported to Excel or Word.

AnalysisThe question it answersUse it when
Cronbach's AlphaDo my items hang together as one scale?You already have a scale and want to report its reliability.
PCACan I boil many variables down to a few summary scores?You want to simplify data or combine overlapping variables.
EFAWhat hidden ideas are behind my items?You are developing or exploring a questionnaire and do not yet know its structure.
To confirm a structure you already decided on (rather than discover one), use Confirmatory Factor Analysis in Structural Equation Modelling.

Cronbach's Alpha

ReliabilityMulti-item scale
In plain words. Alpha tells you whether the questions in a scale agree with each other. If five questions are all meant to measure stress, people who score high on one should tend to score high on the others. Alpha summarises that agreement as a number from 0 to 1. Think of five bathroom scales: if they all give roughly the same reading for the same person, you can trust the average. If they disagree wildly, you cannot.
Use it when
  • You use a set of questions that you will add or average into one score
  • A funder, journal or ethics board wants evidence your scale is reliable
Your data needs
  • Numeric items (Likert answers work well)
  • Items that all point the same way. Reverse-worded items must be flagged first.
  • At least 3 items, and ideally 100 or more respondents

Cronbach's Alpha: how to run it

1
Select the scale items
Choose the numeric variables from the multiselect. All should measure the same idea.
2
Click Check items for reverse coding
AnalyZ looks at the inter-item correlations and flags any item that correlates negatively with the majority. These are likely reverse-worded (for example I feel calm among stress items) and should be ticked before you run.
3
Click Run Cronbach's Alpha
You get the overall alpha, the inter-item correlation matrix and the item-total statistics.

Cronbach's Alpha: reading the results

You seeIt means
Cronbach's alphaOverall reliability. See the table below.
Standardised alphaThe same, if all items were put on the same scale first. Useful when items use different response ranges.
Inter-item correlation matrixHow strongly each pair of items agrees. The average is ideally 0.2 to 0.5. Much higher means items are near-duplicates, and negative values point to a reverse-coded item.
Corrected item-total correlationHow strongly each item agrees with the rest of the scale. Aim for 0.30 or higher. A low value means the item does not belong.
Alpha if item deletedWhat alpha would be if that item were removed. If removing one item raises alpha substantially, consider dropping it.
AlphaInterpretation
0.90 or higherExcellent (above about 0.95 can mean items are repeating each other)
0.80 to 0.89Good
0.70 to 0.79Acceptable
0.60 to 0.69Questionable
Below 0.60Poor. Consider revising the scale
Worked example

The situation. A 5-item stress scale, answered on a 1 to 5 scale by 150 people.

ItemItem-total rAlpha if deleted
Stress10.560.742
Stress20.510.759
Stress30.580.734
Stress40.600.729
Stress50.550.747

What AnalyZ shows. Cronbach's alpha = 0.78. Average inter-item correlation about 0.54.

How to read it. An alpha of 0.78 falls in the acceptable-to-good band. Every item correlates 0.30 or more with the rest, and removing any item would lower alpha (every alpha-if-deleted figure is below 0.78), so there is no reason to drop any. The scale is reliable enough to add the five answers into one stress score.

Illustrative figures.

Cronbach's Alpha: things to watch for

Cronbach's Alpha: key terms

Internal consistency
How well items agree with each other.
Reverse-coded item
A question worded in the opposite direction, so its scores must be flipped.
Item-total correlation
How closely an item tracks the rest of the scale.

Principal Component Analysis (PCA)

Simplify dataStandardised by default
In plain words. PCA squeezes many overlapping variables into a few components, each a weighted mix of the originals. The first component captures the biggest shared pattern, the second the biggest remaining pattern, and so on. It is a way of summarising, much as a few headline figures summarise a long report. It makes no claim about hidden causes.
Use it when
  • You want fewer variables for a regression, or want to remove overlap between predictors
  • You want a single score from several related variables such as an asset index
Your data needs
  • Several numeric variables that are correlated
  • At least 5 cases per variable, ideally many more

Principal Component Analysis: how to run it

1
Choose the variables
Variables are standardised by default so large-scale variables do not dominate. Untick Standardise variables to use raw scales.
2
Choose how many components to keep
The default is eigenvalues above 1, but look at the scree plot too.
3
Click Run PCA
Read KMO first, then the variance explained, then the loadings.

Principal Component Analysis: reading the results

You seeIt means
KMO measureWhether your variables are correlated enough to summarise. 0.60 or higher is acceptable. Below that, PCA may not be worthwhile.
Eigenvalues and variance explainedHow much of the total variation each component captures. An eigenvalue above 1 means the component explains more than one original variable would on its own.
Scree plotA line chart of eigenvalues. Keep components up to the 'elbow', where the line flattens.
Component loadingsHow strongly each variable relates to each component. Above 0.4 (or below -0.4) is meaningful. Look at which variables load highly to name the component.
CommunalitiesHow much of each variable the kept components capture. Below 0.40 is flagged as poorly represented.
Worked example

The situation. Eight household asset questions (radio, phone, bicycle, and so on) from 500 households.

What AnalyZ shows. KMO = 0.82. Component 1: eigenvalue 3.4, 42% of variance. Component 2: eigenvalue 1.3, 16%. Component 3: eigenvalue 0.8.

How to read it. KMO of 0.82 means the variables are well suited to summarising. Two components pass the eigenvalue rule and together explain 58% of the variation. If every asset loads positively on Component 1, it can serve as a general wealth index. Component 2 might separate productive assets (such as a plough) from consumer goods. Any variable with communality below 0.40 is not well captured, and you might drop it and re-run.

Illustrative figures.

Principal Component Analysis: things to watch for

Principal Component Analysis: key terms

Component
A weighted combination of your original variables.
Eigenvalue
The amount of variation a component captures.
Communality
How much of a variable the retained components explain.
KMO
A check that your data suit factor-style methods.

Exploratory Factor Analysis (EFA)

Discover structureQuestionnaire development
In plain words. EFA assumes your answers are caused by a smaller number of unseen ideas (factors). A stress questionnaire might turn out to contain two ideas, worry and physical tension. EFA finds those groupings by looking at which questions rise and fall together. Unlike PCA, it is a model of hidden causes, not just a summary.
Use it when
  • You are developing or shortening a questionnaire
  • You do not know how many dimensions your questions contain
Your data needs
  • Numeric items that are correlated
  • At least 100 cases, and about 5 to 10 cases per item

Exploratory Factor Analysis: how to run it

1
Choose the items
Include only items you think relate to the topic.
2
Choose the number of factors
Guided by eigenvalues above 1, the scree plot elbow, or parallel analysis.
3
Choose extraction and rotation
Extraction: Principal Axis Factoring (the usual choice for finding hidden factors) or Principal Components. Rotation: Varimax (factors independent), Oblimin (factors may correlate) or None.
4
Click Run EFA
Read KMO, then the variance explained, then the rotated loadings.

Exploratory Factor Analysis: reading the results

You seeIt means
KMO0.60 or higher is acceptable.
Eigenvalues, variance explained and scree plotAs in PCA.
Factor loadingsHow strongly each item reflects each factor. Above 0.4 suggests a good indicator. An item that loads strongly on two factors (a cross-loading) is ambiguous and may need rewording or removal.
Factor correlation matrix (oblique rotation)How related the factors are. If all correlations are below 0.32, Varimax would have given about the same answer.
CommunalitiesBelow 0.30 is flagged as poorly explained by the factors.
Worked example

The situation. Ten wellbeing questions from 300 people, two factors extracted with Principal Axis Factoring and Oblimin rotation.

What AnalyZ shows. KMO = 0.86. Two factors explain 54% of the variation. Factor correlation = 0.41.

How to read it. Items 1 to 5 (all about mood) load 0.60 to 0.80 on Factor 1 and almost nothing on Factor 2. Items 6 to 10 (all about sleep and energy) do the opposite. That reveals two ideas, emotional wellbeing and physical wellbeing. One item loading 0.45 on both would be a candidate for rewording. The factors correlate at 0.41, above 0.32, so Oblimin was the right choice. Once the structure is found, confirm it on a fresh sample with CFA.

Illustrative figures.

Exploratory Factor Analysis: things to watch for

Exploratory Factor Analysis: key terms

Latent factor
An unseen idea that causes related answers.
Rotation
A step that makes the factor pattern easier to read.
Cross-loading
An item that relates strongly to more than one factor.

Survey weights

When a survey weight is set in Data View, all three analyses use it by default. Untick Apply survey weight beside Run to compare with the unweighted result. Each analysis works from the survey-weighted covariance or correlation matrix of your items, so respondents count in proportion to their weights, and Weighted N appears next to N.

AnalysisWhat changes
Cronbach's AlphaAlpha, standardised alpha and the inter-item correlations come from the weighted matrix. Item means and standard deviations are weighted, and the corrected item-total correlations and alpha-if-item-deleted use the weighted covariances. Reverse-coded items are detected from the weighted correlations.
PCAEigenvalues, variance explained, loadings, communalities and KMO come from the weighted correlation matrix, or from the weighted covariance matrix if you do not standardise.
EFAExtraction and rotation work on the weighted correlation matrix, so loadings, communalities, variance explained, factor correlations and KMO reflect the weights. Bartlett's test uses the effective sample size implied by the weights, which is smaller than N, so it is more cautious.

Sampling brings together three planning tools that you use before collecting data (or independently of any loaded dataset): Sample Size Estimation, Power Analysis and Draw a Sample. Start on the landing screen, choose the tool, then pick the calculator or design.

Why this matters. A study that is too small will miss real effects and give unreliable results. One that is too large wastes time and money. Most ethics boards and journals ask for a sample size justification.

Five ideas to know before you start

IdeaIn plain words
Confidence levelHow sure you want to be that the true value lies within your margin of error. 95% is standard.
Margin of errorHow far your estimate may be from the truth. A margin of 5% means a survey result of 40% really means 35% to 45%. A smaller margin needs a larger sample.
Statistical powerThe chance of detecting a real effect if it exists. 80% is the usual minimum, 90% for important studies.
Effect sizeHow big a difference you hope to detect. Small differences need much larger samples than big ones. Choose the smallest difference that would matter in practice.
Design effect (DEFF)A multiplier for cluster designs. People in the same village or school resemble each other, so you learn less per person and need more people. Typical field surveys use 1.5 to 2.0.

Sample Size Estimation

Before data collectionNo dataset needed
In plain words. Answers how many people do I need to ask? Calculators are grouped by what your study is doing: describing a population at one point in time (Cross-sectional), detecting a difference or change (Measuring change), or other designs (Case-Control, LQAS).
Use it when
  • You are planning a survey or evaluation
  • A protocol or funder requires a sample size justification
Your data needs
  • Your target confidence level and margin of error (or power and effect size)
  • A rough guess of the proportion or standard deviation you expect, from earlier studies or a pilot

Sample Size Estimation: how to run it

1
Choose the group, then the calculator
Cross-sectional, Measuring change (Longitudinal, Change in Proportion, RCT) or Other designs (Case-Control, LQAS).
2
Enter your parameters
Defaults are shown. Change them to suit your study.
3
Click Calculate sample size
Use Compare confidence levels to see the size at 80, 90, 95 and 99% side by side.
4
Read the interpretation and the formula
The formula is shown at the bottom for citing.

Sample Size Estimation: reading the results

You seeIt means
The main figure (large text)The sample you need, for example per arm in an RCT or total in a cross-sectional survey.
Secondary figuresThe minimum before non-response, number per stratum, or clusters needed.
Interpretation panelPlain guidance and any design-specific caveats.
FormulaThe exact calculation, for transparency and referencing.

Cross-sectional: the flexible calculator

ChoiceOptions
What are you measuring?Proportion (%), a rate or share. Defaults to 50% if you have no estimate, which gives the largest and safest sample. Mean (numeric), an average such as income, which needs an expected standard deviation.
How will you select participants?Simple random, equal chance for everyone. Stratified, subgroups sampled separately (add strata names and sizes, or tick I need this precision within each stratum if you will report by subgroup). Cluster / multi-stage, whole groups first, which also covers PPS designs.
Design effectDefaults to 1 and is always editable. With cluster designs you may calculate it from the intra-class correlation: DEFF = 1 + (m - 1) × ρ, where m is the average cluster size and ρ the ICC. For example, with 20 per cluster and ρ = 0.05, DEFF is 1.95.
Population size and non-responseOptional. A population size applies a finite-population correction. Non-response inflates your recruitment target.

Measuring change and other designs

CalculatorUse whenKey inputs
Longitudinal StudyFollowing the same people over several rounds.Confidence, margin of error, number of waves, dropout per wave, population size
Change in ProportionDetecting a change in a share, between two groups, or in one group before and after.Confidence, power, baseline and endline proportion. For before-and-after, the expected correlation between the two measurements, since the higher it is, the smaller the sample needed.
Randomized Controlled TrialRandom assignment to treatment and control.Confidence, power, control and treatment outcome rates, expected attrition
Case-ControlComparing people with an outcome (cases) against those without (controls).Confidence, power, exposure rate among controls, odds ratio to detect, controls per case
LQASClassifying whether coverage meets a threshold, instead of estimating a precise rate. Common in humanitarian monitoring.Coverage threshold, upper (acceptable) and lower (unacceptable) bounds

LQAS example. With a sample of 19 and a decision rule of 13, a site with truly acceptable coverage has about a 7% chance of being wrongly failed and a site with truly unacceptable coverage has about an 8% chance of being wrongly passed. These are the two error risks the calculator displays so you can judge the rule.

Worked example

The situation. Cross-sectional survey of the percentage of households vaccinated. No prior estimate (use 50%), 95% confidence, 5% margin of error.

StepSample size
Simple random sample (DEFF = 1)385
Clustered design, DEFF = 2770
Add 10% expected non-response855
Simple random sample if the population is only 2,000 (finite-population correction)about 323

How to read it. A simple random survey needs 385 households. Because you plan to visit villages (clusters), people within them resemble each other, so you must double the sample, to 770. Allowing for 10% of households not responding, plan to approach 855. If the whole population is only 2,000 households, you need fewer (about 323 in the simple random case), because you are sampling a large share of everyone. Other examples: to estimate an average with a standard deviation of 12 to within 2 units needs about 139 people. To detect a rise in a proportion from 40% to 55% with 80% power needs about 170 per group. A case-control study with 30% exposure among controls and an odds ratio of 2 needs about 141 cases (and as many controls at 1:1).

Calculated for illustration.

Sample Size Estimation: things to watch for

Sample Size Estimation: key terms

DEFF
Design effect: the multiplier for clustering.
Finite-population correction
A reduction when you sample a large share of a small population.
Attrition
People lost between rounds.

Power Analysis

Eight testsTwo modes
In plain words. Answers if I use this test, how many people do I need to detect a difference of a given size? or the reverse, given the people I have, what chance do I have of detecting it? Use it for a specific test not covered by the sample size calculators, or to explore several sample sizes.
Use it when
  • You have chosen your analysis and need to justify the sample
  • You have a fixed sample and want to know what you can detect
  • You want to see how power changes across sample sizes
Your data needs
  • The test you plan to use
  • An effect size, drawn from earlier studies or the smallest effect that would matter

Power Analysis: how to run it

1
Choose the test
Two-sample t-test, One-sample t-test, Paired t-test, Chi-square, McNemar, One-way ANOVA, Linear regression or Logistic regression.
2
Choose the mode
Sample size mode: enter a target power to find the required sample. Achieved power mode: enter a sample size to see the power.
3
Enter the effect size, significance level and any other inputs
Small, medium and large benchmarks are shown for each effect size measure.
4
Click Calculate, then Show power curve
The curve shows how power changes with sample size for small, medium and large effects.

Power Analysis: reading the results

You seeIt means
Required sample size or achieved powerThe headline number.
Power curveWhere adding more people stops paying off. The steep part shows where each extra person helps most.
TestUse whenEffect size measure
Two-sample t-testComparing means of two independent groupsCohen's d (0.2 small, 0.5 medium, 0.8 large)
One-sample t-testComparing a group mean with a known valueCohen's d
Paired t-testTwo measurements on the same peopleCohen's dz
Chi-squareAssociation between two categorical variablesCohen's w (0.1, 0.3, 0.5)
McNemar's testChange in a yes/no outcome in the same peopleDiscordant pair proportions
One-way ANOVAMeans across 3 or more groupsCohen's f (0.1, 0.25, 0.4)
Linear regressionDetecting a given R² with several predictorsCohen's f² (0.02, 0.15, 0.35)
Logistic regressionBinary outcome with several predictorsOdds ratio and events per variable (EPV)

Use the paired t-test or McNemar's test when the same people are measured twice, and the two-sample t-test or chi-square when comparing separate groups. Paired designs generally need fewer people for the same effect.

Worked example

The situation. Planning to compare two groups with a t-test, 80% power, 5% significance level.

Test and effect sizeResult
Two-sample t-test, d = 0.5 (medium)64 per group
Two-sample t-test, d = 0.3 (small to medium)176 per group
One-sample t-test, d = 0.534 in total
One-way ANOVA, f = 0.25, 3 groupsabout 158 in total (about 53 per group)
Chi-square, w = 0.3, 1 degree of freedom88 in total
Linear regression, f² = 0.15, 5 predictors92 in total

What AnalyZ shows. Achieved power with d = 0.5 and 30 per group: 0.48.

How to read it. Halving the effect size (0.5 to 0.3) nearly triples the sample needed, which is why choosing a realistic effect size is the most important step. If you can only recruit 30 per group, you have a 48% chance of detecting a medium effect, worse than a coin flip. Either recruit more, accept that you can only detect larger effects, or use a design with more power, such as a paired comparison.

Calculated for illustration.

Power Analysis: things to watch for

Power Analysis: key terms

Power
The chance of detecting a real effect.
Effect size
How big the difference is, in standardised units.
Significance level
The risk of a false alarm you accept, usually 5%.

Draw a Sample

Needs a sampling frameFive designsReproducible
In plain words. Once you know how many, this tool picks who. You upload a list of every unit in your population (the sampling frame, such as all households or all schools), and AnalyZ draws the sample for you using a proper random method.
Use it when
  • You have a full list of the population and need a defensible random selection
  • You need the same draw to be repeatable later
Your data needs
  • A CSV, XLS or XLSX file with one row per unit
  • Columns for strata, clusters or size, depending on the design

Draw a Sample: how to run it

1
Upload your frame
AnalyZ shows the number of rows and columns once loaded.
2
Choose a design
See the table below.
3
Set the sample size and the design settings
A strata column and allocation method for stratified. A cluster column and number of clusters for cluster or PPS. A size column for PPS.
4
Optionally set a random seed
A seed makes the draw repeatable. Write it in your protocol.
5
Click Draw sample
Review the preview table, then download as CSV or Excel, either the selected rows alone or the full frame with a _selected flag column.

Draw a Sample: reading the results

You seeIt means
Preview tableThe selected units. Check it looks sensible before downloading.
Download optionsSelected rows only, or the whole frame flagged.
DesignUse when
Simple RandomEach unit has an equal chance of selection. The default when there is no further structure.
SystematicTaking every k-th unit after a random start. The frame should be in a random order.
StratifiedYou must guarantee representation of subgroups such as region or sex.
ClusterYou select whole groups (villages, schools) and survey everyone, or a sample, within them.
PPS (probability proportional to size)Multi-stage designs where clusters differ in size. Larger clusters have proportionally higher selection chances.
Worked example

The situation. A list of 1,200 households across 3 districts. You need 300, with enough from each district.

How to read it. A simple random draw could, by bad luck, take very few from the smallest district. A stratified draw with district as the strata column guarantees each district gets its share. Setting a seed (say 2025) means a colleague can regenerate the exact same list from the same file later.

Draw a Sample: things to watch for

Draw a Sample: key terms

Sampling frame
The complete list the sample is drawn from.
Stratum
A subgroup sampled separately.
PPS
Selection chance proportional to size.

The Qualitative module is organised into five sub-modules. Load your transcripts first (see Loading Data → Qualitative data), then open Qualitative Analysis from the home screen. Use the buttons below to jump directly to any sub-module's guide.

View Transcripts
Read, edit, and tag transcripts. Overview stats, word cloud, word search, and coding stripes.
Thematic Coding
Build a codebook, assign and remove codes, auto-suggest with keywords, and query coded segments.
Compare Coding
Load two coders' exports and compute Cohen's Kappa for inter-rater reliability.
Explore Themes
Thematic saturation, theme comparison across groups, theme connectedness, and segment length.
Analyze Relationships
Network Map, Mind Map, Thematic Clusters, Code Sequence, Configurational Analysis, QCA, and Persona Builder.

🔄 Recommended workflow

Upload transcripts
→
View & clean
→
Build codebook
→
Assign codes
→
Compare coding
→
Explore themes
→
Analyze relationships
→
Save session
Closing the browser without saving will lose all coded segments, unless you've turned on Save on this device (and even then, it expires after 24 hours and is only restored on page refresh, not after fully closing the browser). Refreshing the same tab is safe. Save your session using the Save Session button in the sidebar for anything you want to keep, this saves all transcripts, your full codebook (codes, memos, keywords, colours), and every coded segment.

View Transcripts

Opens with a card selection screen. Four tabs are available: Transcripts, Word Cloud, Word Search, and Transcript Settings. Use the Back button to return to the card selection at any time.

Transcripts, overview and detail

The Transcripts tab opens in overview mode: all loaded transcripts displayed as cards showing word count, sentence count, average sentence length, and coded segments. Click any card to open that transcript's detail view.

In detail view: read the transcript in full, with coding stripes shown in a thin column on the right margin, coloured bars indicating which codes are applied at each point in the text. A "What the stripes mean" key below maps each colour to its code. Click Edit to make corrections, Save to commit, or Cancel to discard. Inline tag dropdowns let you assign participant attributes without leaving the transcript view. Click All transcripts to return to the overview.

The coding stripes give you a quick visual sense of coding coverage, long dark stretches with no stripes may indicate uncoded passages worth revisiting in Thematic Coding.

Word Cloud

Shows word frequency across all transcripts, a single transcript, or coded segments only. Adjust the max words slider, exclude specific words using the exclusion list with autocomplete suggestions, and filter by a specific code when in coded segments mode. The frequency table below lets you exclude words individually. Export as CSV.

Word Search

Search for any word or phrase across all transcripts or a single transcript. Each result shows the matching sentence in context with the search term highlighted. Toggle case sensitivity as needed. Export all matches as CSV.

Transcript Settings, tagging transcripts

Define tag categories to classify each transcript by participant attributes. This enables grouped comparisons in Explore Themes.

1
Add a tag category
Click Add category. In the first box write the category name (e.g. Gender). In the second box write the values separated by commas (e.g. Male, Female, Non-binary). Click the tick button to save, or press Enter.
2
Assign values to each transcript
In the Assign Tags to Transcripts panel, use the dropdowns to select the appropriate value for each transcript in each category. Tags can also be assigned from within the transcript detail view.
3
Review the tag summary
A summary table shows all transcripts and their assigned tags. Categories can be deleted, this removes them from all transcript tags too.

Thematic Coding

What it does: Assigns meaning to sections of your transcripts by tagging them with codes, labels that capture what is happening in that passage. Over time, a pattern of codes builds into themes that describe recurring ideas, experiences, or perspectives in your data.

What you will see: A card selection screen with three cards: Manage Codes, Assign Codes, and Review Codes. Each card opens its own sub-view; the Back button in the header always returns you to the card selection.

Technical note: AnalyZ supports inductive coding (codes emerge from the data, you read first and define codes as you go) and deductive coding (codes are defined in advance from a framework or theory). Most qualitative researchers use a mix of both. The codebook approach in AnalyZ is consistent with Braun and Clarke (2006) thematic analysis and framework analysis. A single text passage can be assigned multiple codes, and the same code can appear across different transcripts, both are intentional features that support nuanced, overlapping themes.

Thematic Coding opens with a card selection screen, Manage Codes (your codebook), Assign Codes (the coding workspace), and Review Codes (audit all coded segments). Click a card to enter that section. Use the Back button to return to the card selection at any time.

Manage Codes, building your codebook

Click ➕ Add new code. Enter a code name (e.g. Housing Insecurity), choose a colour, and optionally add a description. Click Add code. Each code shows a usage count (how many segments it has been assigned to). Codes can be edited (✏️ button, opens an inline form) or deleted (🗑 button, also removes the code from all segments).

Start with broad codes and refine them as you read more transcripts. This inductive approach is standard in thematic analysis (Braun & Clarke, 2006) and grounded theory. You can also add new codes on-the-fly from within the Assign Codes screen.

Manage Codes, building the codebook

Create, edit, colour, and delete codes. Each code can have a name, description, analytical memo (a working note on when and how to apply it), and keywords for auto-suggest. The colour picker shows 15 preset swatches; click + for a custom colour.

Merging codes. Open a code with the edit button, choose Merge into another code and click Merge. Every segment carrying the old code moves to the code you chose, and the old code is removed. Use it to combine near-duplicates as your codebook settles. Renaming a code updates all of its existing segments automatically.

Assign Codes, the coding workflow

The screen is split: transcript on the left, Coding Panel on the right.

1
Select a transcript
Use the dropdown at the top of the screen to choose which transcript to code.
2
Select a text passage
Click and drag in the transcript panel to highlight the passage you want to code.
3
Click a code to assign it
With text selected, click any code button in the Coding Panel. The excerpt is saved immediately. Coded passages are marked with a coloured underline.
4
Assign multiple codes or remove one
Click additional codes to assign them to the same passage. Already-assigned codes show a ✓ prefix and a small ✕ badge at the top-right corner, click the badge to remove just that code from the segment without affecting others.
5
Use keyword auto-suggest
Click Auto-assign codes in the toolbar. AnalyZ scans the transcript for keywords defined in each code's codebook entry and proposes assignments. Review each suggestion with Accept or Skip, or use Accept all.
6
Save your session
After coding, a reminder nudge appears suggesting you save from the sidebar. Use Save Session to download a .analyz.json file preserving all transcripts, codes, and coded segments.
More in Assign Codes. Use the search bar to find passages by word, and switch on Show uncoded only to focus on lines you have not coded yet. Before clicking a code, you can type an analyst note to record why you coded a passage that way. Notes appear in Review Codes and in CSV exports.

Review Codes, auditing and querying segments

Review Codes opens with three cards. Click any card to enter that view. Use the Back button in the top header to return to the card selection.

  • Coded Segments: browse, filter, sort, edit, and export all coded segments. Two filter modes: Simple filter (transcript and code dropdowns) and Query builder (Boolean AND/OR/NOT conditions across codes). Export as CSV, Excel, or JSON (for inter-rater reliability).
  • Code Frequency: frequency table, bar chart, and Code × Transcript heatmap showing how often each code appears across transcripts. Includes a plain-language interpretation panel.
  • Coding Map: heatmap showing where each code appears within a single transcript, by paragraph or by interview quarter. Coverage bars show what proportion of the transcript each code covers. Use this to spot uncoded stretches and check coding coverage across the interview.

Compare Coding

A standalone module for inter-rater reliability analysis. Use it when two researchers have independently coded the same transcripts and want to measure agreement.

1
Export from both coders
Each coder uses the JSON (IRR) button in Review Codes to download their coding as a JSON file. This must be done before opening Compare Coding.
2
Load both exports
In Compare Coding, load each coder's JSON file and give them a label (e.g. Coder A, Coder B). Click Compute agreement score.
3
Read the results
The overall Cohen's Kappa (macro average) is shown alongside per-code Kappa values and percentage agreement. The interpretation panel categorises each code as strong (κ ≥ 0.8), acceptable (0.6–0.8), or below threshold (<0.6), with specific next steps for low-agreement codes.
Cohen's Kappa corrects for chance agreement, it is a more conservative measure than simple percentage agreement. κ ≥ 0.6 is the widely accepted minimum for qualitative research. For codes below threshold: compare segment lists side by side, revisit the code definition and memo, run a calibration session, then re-export and recompute.

Explore Themes

Requires coded segments to be present. Opens with four cards in a 2×2 grid. Each card opens its own analysis view with a plain-language interpretation panel. Use the Back button to return to the card selection.

CardWhat it showsBest for
SaturationBar and line chart showing new codes introduced per transcript (cumulative codebook growth)Assessing whether coding has reached thematic saturation, a flat line means no new themes are emerging
Theme ComparisonTwo modes: Per transcript (code share heatmap and raw counts table) and By participant group (normalised frequency by tag category, bar chart and heatmap)Per transcript: seeing which themes dominated each interview. By group: comparing theme prevalence across subgroups (e.g. gender, location, domain)
Theme ConnectednessTwo views toggled within the card: Co-occurrence (heatmap and ranked table of how often code pairs appear together) and Distinctiveness (exclusivity scores for code pairs that rarely or never appear together)Co-occurrence: finding theme clusters and candidates for higher-level labels. Distinctiveness: identifying themes that occupy separate conceptual territory
Segment LengthAverage word count per coded segment for each code, sorted by elaboration depthIdentifying which themes participants elaborate on at length, a proxy for depth of engagement or emotional significance
By participant group in Theme Comparison requires tag categories set up in View Transcripts → Transcript Settings first. Values are normalised as a percentage of total code assignments per group, so comparisons are fair regardless of interview length. Add to Canvas and Add to Sense Making buttons are available on the heatmap and summary table sections of each view.
Technical note, Saturation: Theoretical saturation is the point at which adding more transcripts no longer introduces new codes. The saturation curve plots cumulative unique codes as transcripts are added in the order they appear in your file list. A flattening curve suggests saturation, though this depends on the richness and diversity of your data, not just transcript count. Homogeneous samples often saturate at 12–15 interviews; heterogeneous samples may require 20–30 or more.

Analyze Relationships

Explores structural and configurational relationships between codes and participants across your corpus. Requires at least 2 codes and 2 coded transcripts. Opens with seven cards. Each card includes an automated plain-language interpretation panel and Add to Sense Making / Add to Canvas buttons.

CardWhat it showsHow to read it
Network MapDraggable SVG network where codes are nodes (sized by frequency) and edges represent co-occurrence counts. Toggle connection count labels and adjust edge colour. Export as image for Canvas.Hub codes connected to many others are likely core themes. Isolated codes may be peripheral or under-coded. Drag nodes to explore the structure.
Mind MapA branching diagram with your central topic in the middle. Each branch is a group of codes that tend to be applied to the same segments (or to the same transcripts), and the number beside a code is how many segments carry it. Controls: Centre (all codes, or one code to see everything linked to it), Central topic (the label in the middle), Group codes that share a Segment or Transcript, Branches (Auto or a number), Min segments per code, Show segment count, and Expand all / Collapse all. Click a branch to fold it, and hover over a code to read an example segment. Add to Canvas saves the map as an image slide.Branches are suggestions from your data, not finished themes: read the codes on a branch together and name the theme yourself. Choose a single code under Centre to list the codes that share a segment with it, strongest link first. Codes that share nothing with any other appear on a grey Unlinked codes branch. If every code lands on one branch, group by Transcript, or add a second code to segments that cover more than one theme.
Thematic ClustersJaccard similarity matrix and presence heatmap, codes that appear in the same transcripts are grouped togetherTight clusters of co-occurring codes are candidates for higher-level theme labels. Codes that cluster together across transcripts form the backbone of your thematic structure.
Code SequenceSankey diagram of code-to-code transitions, the order themes follow one another within transcripts. Self-transitions (same code consecutive) are filtered out.Dominant transitions reveal narrative structure and causal reasoning. A theme that consistently precedes another may be a precondition or trigger for it.
Configurational AnalysisPresence/absence matrix of codes per transcript, sorted so transcripts with identical thematic profiles appear together. Identifies participant types (configurations) and classifies codes as Universal, Differentiating, or Rare.Transcripts sharing the same configuration form a participant type. Universal codes are present in every transcript. Differentiating codes are what distinguish one type from another. Rare codes appear in only one transcript and should be treated as individual cases rather than group patterns.
Qualitative Comparative AnalysisCrisp-set QCA in four steps: select conditions (codes), assign a binary outcome per transcript, build a truth table with consistency scores, and compute the minimised Boolean solution. Consistency threshold is adjustable (default 0.75). Add to Sense Making sends the full truth table and solution. Add to Canvas sends the truth table as a slide.Use when you want to explain why some transcripts achieve an outcome and others do not. Identifies the minimal combinations of coded themes sufficient for the outcome. Requires at least 2 coded transcripts with outcomes assigned and at least 2 conditions selected. Outcome assignments are session-only and do not affect other analyses.
Persona BuilderIdentifies thematic personas from the co-occurrence patterns in coded segments in three steps. Step 1: select which codes define the thematic space. Step 2: review the silhouette-scored suggestion, the analysis tests cluster solutions from 2 to 8 personas and recommends the most analytically distinct solution; click any row to select a different number. Step 3: persona cards appear in a grid with auto-generated descriptive names, thematic frequency profiles, participant characteristics from transcript tags, and sample segments. Expand a card to see frequency bars and click a theme to view all segments. Use Merge into to combine similar personas. Add to Sense Making sends all personas with frequency data and interpretation. Add to Canvas sends a frequency profile table.Suitable for both IDIs and FGDs. Personas represent thematic positions, not individual participants or transcripts. A silhouette score above 0.7 indicates strong separation between personas; below 0.5 suggests the code selection should be reconsidered. Participant characteristics (Predominantly female, Mostly rural) are computed from transcript tags counted per unique source file.
Use the Min codes per transcript slider in Configurational Analysis to filter out lightly coded transcripts. Set it to 2 or higher to focus the typology on transcripts where coding is substantive. The code role legend (Universal / Differentiating / Rare) updates as you adjust the slider. For QCA, select conditions before assigning outcomes and set the consistency threshold before viewing the truth table. For Persona Builder, select codes that vary meaningfully across participants, universal codes (present everywhere) and very rare codes add little discriminating power.
Technical note, Code clustering: AnalyZ clusters codes using hierarchical clustering on a co-occurrence distance matrix. Codes that frequently appear together are grouped first.

Technical note, Configurational Analysis: Each unique combination of present/absent codes across transcripts constitutes a configuration. Configurations shared by 2 or more transcripts are analytically robust; singleton configurations (unique cases) may represent outliers.

Technical note, Persona Builder: Personas are derived from hierarchical clustering of coded segments using Jaccard similarity on code co-occurrence patterns. The silhouette score measures how well-separated the clusters are (above 0.7 is strong, 0.5 to 0.7 is reasonable, below 0.5 suggests overlap). Personas represent thematic positions in the data, not individual participants, and are suitable for both IDI and FGD transcripts without requiring speaker attribution.

Technical note, QCA: AnalyZ implements crisp-set QCA (csQCA) using the Quine-McCluskey algorithm for Boolean minimisation. The conservative solution approach is used, logical remainders (unobserved configurations) are excluded from minimisation. Outcome assignments are stored in session state only and are not written to transcript data, so the same transcripts can be used for multiple QCA runs with different outcome variables.

Key ideas in plain language

IdeaIn plain wordsExample
CodeA short label you attach to a piece of text to say what it is about.'Cost of transport' on a sentence about the bus fare to the clinic.
CodebookThe list of codes with a definition and when to use each, so everyone codes the same way.A shared table: code name, meaning, include and exclude rules.
ThemeA bigger pattern formed by several related codes.The codes bus fare, distance and time off work together form the theme barriers to access.
SaturationThe point where new interviews stop producing new codes or ideas.If the last three interviews added no new codes, you may have enough data on that topic.
Inter-coder agreement (Cohen's kappa)How often two coders agree, after removing the agreement expected by chance. Above 0.60 is substantial and above 0.80 is strong.Two researchers code the same 20 segments. Kappa of 0.72 means good agreement. Discuss the disagreements and refine the codebook.
Start with a few interviews, build a draft codebook, then test it with a second coder before coding everything. Changing definitions early is cheap, and changing them after you have coded all transcripts is not.

Sense Making brings together the analyses you have run across any module, quantitative tests, qualitative codes, visualisations, and uses AI to synthesise them into a single coherent interpretation. Instead of describing each result individually, the AI identifies patterns, connections, and contradictions across your full body of evidence.

Free plan limit: 1 interpretation per calendar month (each refinement also counts as one). Starter plan: 10/month. Pro plan: 50/month. Your remaining allowance is shown in the project workspace.

How it works

1
Run your analyses
Work normally across any modules, Descriptive, Inferential, Qualitative, Visualization, Scale & Reliability. There is no special mode to activate.
2
Add results to a project
When you see a result worth keeping, click ➕ Add to Sense Making below the export buttons. A dialog opens: choose an existing project from the dropdown, or click + Create a new project to create one inline, no need to navigate to Sense Making first. Edit the title and confirm. If you have no projects yet, the create form opens automatically.
3
Open Sense Making
Open Modules in the top bar, then choose Sense Making under Tools. Your projects are listed with their analysis count.
4
Set up your project
Open a project. Add context (optional but recommended), choose an interpretation mode, and review the list of analyses. You can remove any analysis before interpreting.
5
Interpret
Click Interpret project. The AI synthesis streams in, appearing word by word. When it finishes, a Refine interpretation panel appears where you can instruct the AI to amend the output, the revised version replaces the original.
6
Export
Click ⬇ Export to Word to download the interpretation as a formatted Word document, including context and the full synthesis.

Projects

A project is a named collection of analyses grouped for a specific synthesis. Key facts:

Curate, don't collect. The 20-analysis limit is intentional. The AI produces better interpretations when given a focused, curated set of findings rather than everything you ran. Only add results that are genuinely relevant to the research question.

Interpretation modes

Choose a mode in the project workspace before interpreting. The mode persists and can be changed at any time, switching mode and re-interpreting is one click.

ModeBest forStyle
Plain languageNon-specialist audiences, internal readsSimple words, short sentences, jargon-free
AcademicResearch reports, papers, dissertationsFormal tone, statistical terminology, methodology references
Executive summaryLeadership briefings, decision-makersConcise, action-oriented, implications-first
NarrativeDonor reports, public summariesFlowing prose, story-driven, accessible

Length

Each mode offers a length setting that sets the target word count of the interpretation:

ModeShortStandardDetailed
Plain languageabout 300 wordsabout 600about 1,000
Academicabout 600about 1,200about 2,000
Executive summaryabout 200about 400not offered
Narrativeabout 400about 800about 1,400

Context field

The context field (in the project workspace) tells the AI what this project is about. It is optional but significantly improves interpretation quality. Good context includes:

Context is saved to the project automatically when you click 💾 Save context.

Refining the interpretation

After the initial interpretation, a Refine interpretation panel appears below the synthesis. Type a plain-language instruction and the AI rewrites the interpretation incorporating it, the refined version replaces the original. Examples:

The AI rewrites the full interpretation incorporating your instruction, drawing on the original analyses and any context you provided. The revised version replaces the previous one in the project.

Each refinement counts as one interpretation from your monthly allowance, the same as a new interpretation. Refine as many times as your allowance allows.

Re-interpret and version history

Re-interpret generates a fresh interpretation from your current analyses, optionally steered by a short focus note (for example "lead with the qualitative findings" or "add recommendations"). It replaces the current interpretation, and counts as one interpretation from your monthly allowance. If the response is cut off, nothing is saved or counted and your previous version is kept.

Every interpretation and refinement is stored under Version history in the workspace. Open it to read earlier versions and click Restore to bring one back. Use Copy text to copy the current interpretation to the clipboard, and drag the grip handle on an analysis to change its order in the project.

Exporting the interpretation

Click ⬇ Export to Word in the synthesis view. The downloaded .docx file includes:

Privacy and the AI

The AI receives no raw data. It only sees what you explicitly added to the project: aggregated statistics, auto-generated interpretation text from each module, and any notes you wrote. Your dataset stays in your browser session and is never transmitted to the AI.

Data Canvas turns your analysis outputs into a structured, shareable narrative story, powered by AnalyZense. Unlike the dashboard (which lets people explore data interactively), Data Canvas tells the story of what the data means for a non-technical audience: donors, community members, board members, or anyone who needs to understand findings without doing the analysis themselves.

Open Data Canvas from the home page card or the sidebar navigation.

Best workflow: Add findings to your canvas as you work through your analysis, click Add to Canvas on any result worth including. Then come to Data Canvas, answer three context questions, and let AnalyZense generate the story. Far easier than trying to write headlines and narratives from scratch.

How it works

Data Canvas has two layers:

Adding findings

Every analysis output has an Add to Canvas button. Clicking it opens a modal where you choose the destination canvas (or create one inline). Supported finding types:

TypeSource
ChartAny Plotly chart from any analysis module
TableFrequency tables, crosstabs, regression outputs, test statistics
Infographic / NetworkNetwork map from Analyze Relationships (qualitative)
QuoteCoded participant excerpts from Thematic Coding

Deleting a finding from the Findings tab clears the generated story, you will need to regenerate after removing or adding findings.

Generating the story

1
Collect your findings
Add findings from your analysis modules. Aim for the key results, AnalyZense works better with a curated set than with everything you ran.
2
Click Generate Story
In the Findings tab, click the green Generate Story button. A dialog opens asking three questions.
3
Answer the context questions
What is this study about?: topic, population, geography, purpose. Who is the audience?: their role and organisation. Is there a key message you want to convey?: optional; leave blank and AnalyZense decides. Choose a presentation duration: 5 min (key messages only), 15 min (balanced coverage), or 30 min (full treatment). Your answers are saved and pre-fill the next time you regenerate.
4
Review the generated story
Switch to the Story tab. AnalyZense has grouped your findings, chosen the best layout for each slide, and written plain-language headlines and narratives. Each slide shows the visual alongside the narrative text.
5
Refine individual slides
Click Refine on any slide and give a plain-language instruction, for example: "make the headline shorter", "convert the forest plot to a bar chart", "swap the chart and network positions". AnalyZense updates just that slide. Each refinement counts against your monthly refine allowance.
6
Adjust visual heights
Each visual has a Height − + control below it. Use this to make a chart taller or shorter within the slide. Height adjustments are session-only and do not affect the published version.

Viewing and editing the story

Use the toggle in the canvas header to switch between Vertical scroll (all slides in one scrolling page) and Slide-by-slide (one slide at a time). The choice is saved with the canvas and applies to the published story.

AnalyZense picks one of eight layouts for each slide: single chart, single table, single infographic, chart and table, two charts, chart and quote, chart and text, or full text. To change the wording yourself without using a refinement, click Edit text on a slide and edit the headline and narrative directly.

AnalyZense AI features

ActionWhat it doesFreeStarterPro
Generate StoryReads all findings, decides grouping and layout, writes headlines and narratives for the full story2/month5/month15/month
Refine SlideUpdates a single slide based on a plain-language instruction, headline, narrative, layout, or chart type5/month15/month50/month

Your remaining allowance is shown in the canvas header above the title. Regenerating the story (after adding new findings) counts as one Generate use.

Publishing

1
Set privacy
Use the Private / Public toggle in the canvas header to control visibility. Canvases start private.
2
Click Publish
Click the green Publish button. A dialog opens where you can write an optional note for readers (e.g. version information or study context). Click Publish to make the story live.
3
Copy the link
After publishing, the dialog shows the public URL with a Copy button. You can also click Copy link in the canvas header at any time.
4
Re-publish after changes
If you regenerate or refine slides after publishing, the public link still shows the previous version. Click Re-publish to update it. This lets you make multiple changes and publish them together.

Anyone with the link can read the story in their browser, no AnalyZ account required, no software to install. The viewer shows the story title, each slide with its headline, narrative, and visual, and a navigation bar.

Privacy and the AI

AnalyZense receives only the finding summaries, headlines, context text, chart metadata, and table data. Your raw dataset is never sent to the AI. Participant quotes are sent as written text only, no metadata from your dataset accompanies them.

Use Data Canvas to replace static report attachments. Share the link before a meeting so stakeholders arrive already familiar with the findings, they can scroll through on their phones without any software. For donor reports, a well-structured Data Canvas story is often more readable than a 20-page PDF.

Tables

Every analysis module that produces a results table includes export buttons below the output. The buttons are labelled Excel (.xlsx, several sheets when there are several tables), Word (.docx, ready to paste into a report) and, on some results, Image (a picture of the table).

Charts

In the Visualization module, the HTML button above a chart downloads it as an interactive web page. For a picture, use the camera icon in the chart's own toolbar (top right of the chart, visible when you hover), which saves a PNG. In Descriptive Analysis, the chart menu has Download PNG.

💾 Dataset

Use ⬇ Download Data in the sidebar to export your current (possibly modified) dataset. Available formats:

FormatBest for
JSON (.json)Passing data to other software; no labels embedded
Excel (.xlsx)Easy to open and share; no labels embedded. Reading it back in loses labels.
Stata (.dta)Variable labels and value labels are fully preserved
SPSS (.sav)Variable labels and value labels are fully preserved

💼 Sessions

Use 💾 Save Session in the sidebar to download a complete snapshot of your work as a .analyz.json file, dataset, labels, variable type overrides, connected API sources, dashboard charts, and (for qualitative sessions) all transcripts and coded segments. Click 📂 Restore Session at any time to load that file back in and resume your exact session.