Quantitative Analysis/ Structural Equation Modelling
Following an effect step by step

Path Analysis: How to trace how one thing affects another through the steps in between

Path analysis shows how a set of measured variables affect each other in a chain, such as training leading to knowledge, knowledge to confidence and confidence to profit. This guide covers the ideas, the counting, the calculations, every panel of the output, a worked example and how to report it.

22 min read Structural equation modelling Intermediate

01What is path analysis?

Path analysis maps how a set of variables affect each other, step by step. It is used when one thing leads to another, which leads to a third. A single regression cannot show this, because a regression has only one outcome. Path analysis fits several regressions together and shows how they connect.

The variables are things you measured directly, such as age, a test score or income. If your key variables are scores built from several questionnaire items, use a structural equation model instead.

An example

A livelihoods programme trained 600 small business owners. Each person attended a number of training sessions. You also have a business knowledge score (0 to 100), a confidence score (0 to 100), monthly profit in US dollars, and age. You think training raises knowledge, knowledge raises confidence, and confidence and knowledge both raise profit. You want to know three things. Does each step in the chain hold? How much extra profit goes with one more session? And does the training reach profit only through knowledge and confidence?

What you get out of it

  1. Path estimates
    The size and direction of every arrow, with intervals.
  2. Standardised effects
    The same arrows on a common scale, so strong and weak links can be compared.
  3. Direct, indirect and total effects
    How far one variable reaches another, directly and through the variables in between.
  4. Variance explained
    How much of each outcome the model accounts for.
  5. Fit
    Whether the arrows you left out are consistent with the data.

02When to use it

Use it when you have a clear idea of a chain of influence among variables you measured. It suits programme theories with several steps, such as training, then skills, then practice, then results.

If you only have one step between a cause and an outcome, mediation analysis is simpler. If you have one outcome and many independent predictors, a linear regression is enough. If some variables are scores built from questionnaire items, use a structural equation model.

Path analysis can include yes or no outcomes and ordered outcomes as well as numeric ones. A sample of at least 200 is a sensible minimum, and the more arrows, the more cases you need.

03What you need before you start

Requirements and what happens if they are not met
PointWhat is neededIf it is not met
Measured variablesEvery box in the model is a variable you measured, such as a score, an amount or a yes or no answer.If a key variable is built from several questions, use a structural equation model so the noise in the questions is handled.
Sample sizeAt least 10 cases for every number estimated, and more for models with yes or no variables. The example estimates 12 numbers and uses 600 cases.Estimates and intervals become unstable.
DirectionArrows run one way, from earlier or causal variables to later ones. The data cannot confirm the direction.A wrongly directed arrow can still fit. State your reasoning.
Variable typesNumeric, binary or ordered. Choose the right type for each.The wrong type gives wrong numbers. A yes or no outcome treated as a number gives odd results.
Missing answersCases missing any variable in the model are left out.Check missing values first.

04The idea in a picture

Figure 1 shows the example. Each box is a variable. Each arrow says that the variable at its start helps to explain the variable at its end, after taking the other arrows that lead to that variable into account. The numbers are standardised, so they can be compared.

Figure 1. The chain of effects in the exampleBoxes for training sessions, age, business knowledge, confidence and monthly profit, joined by arrows. Sessions and age lead to knowledge. Knowledge leads to confidence and to profit. Confidence leads to profit. Age also leads to profit.Training sessionsattendedAgeBusinessknowledgeConfidenceMonthly profit(USD)0.61-0.100.490.520.24-0.12
Figure 1. The chain of effects in the example, with standardised path values

Some variables are only causes, such as sessions and age. Some are only results, such as profit. Some are both, such as knowledge and confidence. Variables that are both are what make path analysis different from regression. They are called in-between variables.

Every arrow is a regression. Knowledge is regressed on sessions and age. Confidence is regressed on knowledge. Profit is regressed on confidence, knowledge and age. The software fits all three at the same time.

05How it is calculated

Equations and arrows left out

The model for the example is three equations. Leaving an arrow out is a claim. Here there is no arrow from sessions straight to confidence or to profit, and none from age to confidence. The model says these effects go through the other variables.

That claim can be tested. The software compares the correlations the model predicts with the real ones. If an arrow you left out was needed, the gap shows in the fit. This is why path analysis has a fit test, even though each equation is simple.

Direct, indirect and total effects

The effect of one variable on another along a route is the product of the arrows on that route. If several routes lead to the same place, add them.

route effect = arrow 1 × arrow 2 × ...
indirect effect = sum of all route effects
total effect = direct effect + indirect effect

The effect of sessions on confidence has one route, through knowledge: 1.987 × 0.498 = 0.99. So one more session goes with about 0.99 extra points of confidence.

The effect of sessions on profit has two routes. The first goes through knowledge only: 1.987 × 0.500 = 0.99. The second goes through knowledge and then confidence: 1.987 × 0.498 × 1.045 = 1.04. Adding them gives 2.03. There is no direct arrow from sessions to profit, so the total is the same number.

Intervals for indirect effects

The interval for a product of arrows is not a simple one, because a product of two uncertain numbers is more uncertain than either alone. When all variables are numbers, AnalyZ Solutions uses a standard approximation (the delta method) that works out the spread of the product from the spread of each arrow. When the final outcome is a yes or no variable, the module instead draws many new samples from your data, repeats the analysis on each, and takes the middle 95% of the results. You choose the number of resamples in that case.

06The model in more detail

One equation for each outcome

Each variable that is led to by another gets its own equation. A variable that nothing leads to is a starting variable and has no equation.

knowledge = a × sessions + b × age + error
confidence = c × knowledge + error
profit = d × confidence + e × knowledge + f × age + error

All the equations are estimated together. That matters because a variable such as knowledge is an outcome in one equation and a predictor in another.

Counting what is estimated

The example has 5 variables, so the data contain 5×6/2 = 15 variances and covariances. The model estimates 12 numbers: 6 arrows, 3 error variances for knowledge, confidence and profit, the variances of the 2 starting variables (sessions and age) and their 1 covariance. The difference, 15 - 12 = 3, is the degrees of freedom. The 3 comes from the arrows left out: sessions to confidence, sessions to profit and age to confidence. The fit tests whether leaving those arrows out is consistent with the data. If you draw an arrow between every pair of variables, the degrees of freedom are zero and there is no fit to test.

Relation to ordinary regression

When all variables are numeric and no arrow loops back, each equation gives the same arrow estimates as a separate ordinary regression. What path analysis adds is the whole picture: it multiplies arrows along routes, gives the effects with intervals, and tests the arrows you left out. If your model has only one outcome, path analysis equals a regression.

How the numbers are found

How the estimation depends on your variables
Your variablesMethodIntervals for indirect effects
All numericMaximum likelihood, all equations at onceA standard approximation (the delta method)
Final outcome yes or noOne probit regression for each equationResampling, with a number of resamples and a seed you can set
A mix of yes or no, ordered and numericPolychoric correlations with weighted least squaresFrom the survey-robust or model-based standard errors

A probit regression is like a logistic regression, but it assumes a bell-shaped curve for the underlying tendency. Its numbers are on a different scale, so a probit arrow is not directly an odds ratio. Read its direction and size relative to other arrows in the same model.

Direct, indirect and total effects

indirect effect through one route = product of the arrows on the route
indirect effect = sum over all routes
total effect = direct effect + indirect effect

In the example, sessions reach profit by two routes, 0.99 and 1.03, which add to 2.03 dollars of monthly profit per session. There is no direct arrow from sessions to profit, so the total is also 2.03.

Standardised arrows

A raw arrow is in the units of your variables, such as points of knowledge per session. A standardised arrow is in standard deviations, so you can compare a strong and a weak arrow even if the variables have different units. Use raw arrows to say how much, and standardised arrows to say which is stronger.

Comparing models

To test an arrow, compare the model with and without it. If leaving it out makes the chi-square clearly worse, keep it. AIC and BIC support the comparison: lower is better.

Assumptions

07Reading the output

Model fit

The fit shows whether the arrows you left out fit the data. Read it as in the other models in this module: CFI of 0.95 or higher, RMSEA of 0.06 or lower and SRMR of 0.08 or lower are good. If you include every possible arrow, the model fits perfectly by design, and there is nothing to test.

Path estimates

The table gives each arrow in its raw units, which say how many units the outcome changes for one unit of the cause, and in standardised form. It also gives an interval and a p-value. Raw values tell you what an effect means in practice. Standardised values tell you which links are strong.

Direct, indirect and total effects

For every pair of variables with more than one connection, the module separates the direct, indirect and total effect. It also lists each indirect route on its own, so you can see which one matters most.

Variance explained

For each outcome in the model, this is the share of its differences that the model accounts for.

Types of variables

You say how each variable is treated. Numeric variables are treated as numbers. Yes or no variables and ordered categories can be used as outcomes and predictors. Categorical predictors are turned into comparisons with a reference category, as in regression.

08How to run it in AnalyZ Solutions

Open Structural Equation Modelling and choose Path Analysis. The module is marked Beta.

  1. Add an equation for each outcome
    Pick the outcome, then the variables that lead to it. In the example, knowledge is an outcome of sessions and age. Confidence is an outcome of knowledge. Profit is an outcome of confidence, knowledge and age.
  2. Check how each variable is treated
    Numeric, yes or no, or categorical. Change it if the default is wrong.
  3. Set the number of resamples, if it is offered
    It is used only when the final outcome is yes or no. The default suits exploring, and 1,000 or more is better for a final report.
  4. Click Run Path Analysis
    You get the path estimates, the standardised effects, the direct, indirect and total effects, the variance explained, the fit and a plain-language interpretation.
When your data come from a survey with sampling weights. With a survey weight set, the model is estimated with the weights and the weighted number of cases is shown next to the sample size.

09Reading the output, panel by panel

Every panel in the Path Analysis output
PanelWhat it showsWhat to look for
Model fitChi-square, df, CFI, RMSEA, SRMR, AIC, BICSeveral measures in the good range. If df is zero, no fit is shown.
Path estimatesRaw and standardised arrows, with intervals and p-valuesDirection, size and an interval away from zero.
Direct, indirect and total effectsFor each pair of variables, the three effects with intervalsHow much of an effect goes through the middle steps.
RoutesEach route with its own effectWhich route carries most of the effect.
Variance explainedPercent of each outcome explainedLow values mean other factors matter.
InterpretationA plain-language summaryA first draft. Check it against the tables.

10Interpreting the result

The figures below are invented to show the steps. They come from a real run of the module on a simulated dataset of 600 business owners. Monthly profit averaged about 127 dollars.

Model fit
Chi-squaredfp-valueCFIRMSEASRMR
3.5730.3120.9990.0180.016

Step 1: check the fit. The chi-square is 3.57 on 3 degrees of freedom (p = 0.31), CFI is 0.999 and RMSEA is 0.018. The three arrows left out, sessions to confidence, sessions to profit and age to confidence, are not needed. The chain is a good description of the data.

Path estimates
PathRawStandardised95% CIp-value
Training sessions → Business knowledge1.9870.611.78 to 2.19< 0.001
Age → Business knowledge-0.106-0.10-0.18 to -0.040.003
Business knowledge → Confidence0.4980.490.43 to 0.57< 0.001
Confidence → Monthly profit1.0450.520.91 to 1.18< 0.001
Business knowledge → Monthly profit0.5000.240.36 to 0.64< 0.001
Age → Monthly profit-0.280-0.12-0.41 to -0.15< 0.001

Step 2: read the arrows. Each extra training session goes with 2.0 more points of business knowledge. Each extra knowledge point goes with 0.50 points of confidence. Each confidence point goes with about 1.05 dollars of monthly profit, and each knowledge point with about 0.50 dollars on top of that. Age has small negative links with knowledge and profit. All arrows are clear, since none has an interval that includes zero.

Effects through the chain
EffectDirectIndirectTotal95% CI of indirect
Sessions → Confidence0.000.990.990.82 to 1.16
Sessions → Profit0.002.032.031.68 to 2.38
Knowledge → Profit0.500.521.020.42 to 0.62

Step 3: read the effects. One more session goes with about 2.0 dollars more profit per month (interval 1.7 to 2.4). All of it runs through knowledge, and about half of it also through confidence. Since the model has no direct arrow from sessions to profit, the whole effect is indirect. 1.04 dollars of it runs through knowledge and then confidence, and 0.99 through knowledge alone.

Variance explained
OutcomeExplained by the model
Business knowledge38%
Confidence24%
Monthly profit48%

Step 4: read the variance explained. The model accounts for about 38% of the differences in knowledge, 24% in confidence and 48% in profit. A large part of each remains unexplained, which is normal.

In plain words

Business owners who attended more sessions knew more about running a business. Those who knew more felt more confident, and those who knew more and felt more confident earned more. The training appears to raise profit by building knowledge and confidence, with no sign of a separate route. The model describes patterns in this sample. Owners who chose to attend more sessions may differ in ways that were not measured, so this is not proof that training causes the gain.

11How to report the result

Example write-up. A path analysis of 600 business owners estimated the links from training sessions to business knowledge, confidence and monthly profit, adjusting for age. The model fit well (chi-square = 3.6, df = 3; CFI = 0.999; RMSEA = 0.018; SRMR = 0.016). Each extra session was linked to 1.99 more points of knowledge. Through knowledge and confidence, each session was linked to about 2.03 dollars more monthly profit (95% interval in the table), split almost evenly between the route through knowledge alone (0.99) and the route through knowledge and then confidence (1.03). Because these are survey data, the results show associations consistent with the model and do not by themselves show cause.

12A checklist before you trust the result

  1. Every arrow has a reason in theory or design.
  2. Variable types are correct.
  3. You have at least 10 cases per estimated number.
  4. Degrees of freedom are above zero, so the fit can be tested.
  5. Fit is good on several measures.
  6. You read indirect and total effects from the effects panel.
  7. You have named important variables that are not in the model.
  8. Your wording says "linked to" unless the design supports "caused".

13Common mistakes

Treating the arrows as proven causes. They are your theory. The data can show whether the theory fits, not that it is true.

Ignoring the order in time. A cause has to come before its result. If variables were all measured at the same moment, say that in your report.

Leaving out an important variable. If something affects both the cause and the result and you did not measure it, the arrow between them can be misleading.

Adding every possible arrow. The model then fits by design and tests nothing.

Reading standardised values as amounts. Use raw values to say what an effect means in dollars, points or days.

Too many arrows for the sample. Simplify when the model will not run or gives odd values.

Frequently Asked Questions
What is the difference between path analysis and regression?
Regression has one outcome. Path analysis has several outcomes at once, and an outcome in one equation can be a cause in another. That lets it show chains of effects.
What is the difference between path analysis and a structural equation model?
Path analysis uses variables you measured directly. A structural equation model also includes topics measured by several questions and corrects for the noise in them.
Can I include a yes or no outcome?
Yes. Binary and ordered outcomes can be used. The module then uses a method made for those kinds of answers.
What does it mean that my model has no fit test?
If every possible arrow is included, the model fits perfectly by design, so there is nothing to test. Leave out arrows that your theory says should not exist, and the fit becomes informative.
How many resamples should I use for the intervals?
Resampling is used only when the final outcome is yes or no. The default works for exploring. Use 1,000 or more for a final report, because intervals become steadier. With all-numeric variables no resampling is needed.
Can I use a variable like sex or region?
Yes. Add it as a predictor. Categorical predictors are compared with a reference category you choose.
How is path analysis different from several regressions?
For numeric variables each arrow equals a regression estimate. Path analysis adds the whole-model fit, effects through routes with intervals, and tests of arrows you left out.
Why does my model have zero degrees of freedom?
Every possible arrow is in the model. Leave out arrows your theory says should not exist to get a testable model.
Can I use a variable that is a score from several questions?
You can, but the score contains noise that weakens its arrows. A structural equation model handles this better.
What is a probit arrow?
The arrow in a probit regression for a yes or no outcome. Read direction and relative size. It is not an odds ratio.
Does a good fit mean the arrows are causal?
No. It means the pattern of the data is consistent with your model. Other models may fit as well.

Map a chain of effects

Open Structural Equation Modelling in AnalyZ Solutions. Free to sign up, free to analyze.

Get started
Related Guides