1. How do I test for mediation in SPSS when I have control variables? 2. How do I perform linear regression analysis with several control variables

To test for mediation in SPSS with control variables and perform linear regression analysis with multiple control variables, follow these steps:

Testing for Mediation in SPSS with Control Variables:

Performing Linear Regression Analysis with Control Variables in SPSS:

Thanks Muslim Khan

However I need some more guidance/clarifications please.

For my first question when I click on Statistics what comes up is Linear Regression: Statistics. There is no resampling under it rather I see options like Confidence Interval, Covariance Matrix and Collinearity Diagnostics etc. I have a Bootstrap option further down. Should I check the option Perform bootstrapping? The other options there are Sampling (simple and stratified)

For my second question after entering my DV in the dependent box and my IV in the independent box (i.e. Block 1 of 1), should I click Next and then place my control variables inside Block 2 of 2? And then click OK?

Uzochi Okonkwo, in addition to the points you noted, the response by MK contains a lot of stuff that just isn't true. E.g., the GUI for REGRESSION does not have a "Covariates" box; nor does the output from REGRESSION show direct and indirect effects. I suspect, therefore, that MK's response was generated by ChatGPT. And in that case, there's no point in asking him for clarification.

Bruce Weaver , https://copyleaks.com/ai-content-detector does estimate Muslim Khan 's response is AI generated, despite this not being stated. You raise one possibility, that he doesn't know the area so could not respond to Uzochi Okonkwo 's query. Given that the answer doesn't address the question and makes obvious errors that anyone could see if they opened SPSS, this possibility may be true. But what if he is does know chatbots often give bad answers (in this case not addressing the actual question), has read through the response, and knows that it is wrong on details (as you mention), but still posts it. Would this be worse? As people are trying to figure out why a handful of people do this on ResearchGate (as far as I can tell it only makes them look bad), it would be good know which of these two possibility, and there may be others, happened in this case.

As far as the actual question, your two controlled variables, I assume these are independent of each other and randomly allocated, and as such exogeneous variables. Do you expect each to be mediate by the same mediator variable(s) and is it the same outcome variable?

I faced the same issue and this team solved my issue for free : www.Stats4Edu.com

1- Testing for mediation in SPSS when you have control variables involves conducting a series of regression analyses. The most commonly used method for testing mediation is the Baron and Kenny approach, which involves three steps:

1. Establish the relationship between the independent variable (IV) and the mediator (M):

- Run a regression analysis with the IV as the predictor and the mediator as the outcome variable. This will test the direct relationship between the IV and M.

2. Establish the relationship between the IV and the dependent variable (DV):

- Run a regression analysis with the IV as the predictor and the DV as the outcome variable. This will test the direct relationship between the IV and DV.

3. Test the mediator's role:

- Run a regression analysis with both the IV and the mediator (M) as predictors and the DV as the outcome variable. This will test whether the mediator (M) has a significant effect on the DV, controlling for the IV.

If you find that:

- The IV significantly predicts the mediator (Step 1).

- The IV significantly predicts the DV (Step 2).

- The mediator significantly predicts the DV while controlling for the IV (Step 3).

Then, you have evidence for mediation. However, you may also need to assess the size and significance of the indirect effect (the effect of the IV on the DV through the mediator) to confirm mediation. This can be done using bootstrapping or other methods.

Here's how you can perform these steps in SPSS:

1. Open your dataset in SPSS.

2. Click on "Analyze" in the top menu and select "Regression" and then "Linear."

3. In the Linear Regression dialog box:

- In the "Dependent" field, enter the DV.

- In the "Independent(s)" field, enter the IV and any control variables.

- Click "OK" to run the analysis.

4. Review the output for the regression results, including coefficients, p-values, and R-squared values.

5. Repeat the process for Steps 1, 2, and 3 separately with the mediator, IV, and DV as outcomes, adjusting the predictor variables accordingly.

6. After running all three regressions, analyze the results to determine if there is evidence of mediation as described earlier.

To assess the indirect effect and test for statistical significance, you may need to use SPSS syntax or the PROCESS macro, which is a popular tool for mediation analysis in SPSS. This macro allows you to estimate indirect effects and calculate bootstrap confidence intervals. Be sure to refer to the PROCESS documentation for detailed instructions on how to use it for mediation analysis in SPSS.