I have been testing a multiple parallel mediation analysis by Process Macro (version of 2.16). There are three X variables, six mediators and one outcome in my mediation model. I’m using model 4. Because of that three X variables depend on same mediators and outcome, I want to analyze multiple Xs simultaneously.
In Introduction to Mediation, Moderation, and Conditional Process Analysis (Hayes, 2013) it is recommended that in order to estimate the direct and indirect effects of all k X variables in Figure 6.5, PROCESS must be executed k times, each time putting one Xi in the model as X and the remaining k-1 X variables as covariates. Repeating k-1 times generates the indirect effects for all k X variables. Mathematically, all resulting paths, direct and indirect effects will be the same as if they had all been estimated simultaneously. A common random number seed be used for each run of PROCESS when bootstrapping option, so that the same set of samples will be used when confidence intervals for indirect effects are estimated for different independent variables. In the preceding code, the random number generator is seeded with a common (and arbitrary) seed using the seed= option so that at each run, the bootstrap confidence intervals will be based on the same set of 10,000 resamples from the data.
I executed three times Process with model 4 and each time one X in the model as X and the other two X variables as covariates. But unfortunately, I failed using a common seed number each run. I couldn’t find “seed option or seed number” in the input and output of Process Macro (v2.16).
Could you please help me about that how I can select a common seed at each run in order that the bootstrap confidence intervals will be based on the same set of 5000 resamples from the data?
Thank you for help,
It may not be available "Note that not all features available in the PROCESS macro are available in the custom dialog. For example, there is no option in the dialog for saving the bootstrap estimates, seeding the random number generator, or specifying a specific value of a moderator when conditioning effects." (https://processmacro.org/uploads/3/4/8/5/34858384/dialoginstall.pdf)
process y=dependentvariable/x=X1(one of the your independentvariables)/m=mediator/cov= yourcovariances(if any) X2 X3 ..(otherindependent variables you want to include)/effsize=1/center=1/model=4 /boot=5000/seed=34421.
Then run the command k (the number of your Xs) times, each time include one of the independent variables as X and the others as covariates.
For example for the seconds run, it should be
process y=dependentvariable/x=X2(one of the your independentvariables)/m=mediator/cov= yourcovariances(if any) X1 X3 ..(otherindependent variables you want to include)/effsize=1/center=1/model=4 /boot=5000/seed=34421.
It does not matter what number you used to seed. Note that this is for Process Macro version 3. You should use "vars" command for previous version instead of cov.
The example from (Hayes, 2013)
process vars=y1 iv1 iv2 iv3 med1/y=y1/x=iv1/m=med1/boot=10000/model=4
/seed=5235.
process vars=y1 iv1 iv2 iv3 med1/y=y1/x=iv2/m=med1/boot=10000/model=4
/seed=5235.
process vars=y1 iv1 iv2 iv3 med1/y=y1/x=iv3/m=med1/boot=10000/model=4
/seed=5235.
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running