I intend to run a sequential mediation for a study but am at a lost as to my data. My search shows process macro is not for panel data, but my data is panel. can i be assisted in running a mediation analysis for my panel data?
@Lasisi Olalekan Isiaka. Use the following excel sheet as an example.
Yes you can use panel data to run mediation. There are a lot of studies that used panel data in testing mediation effects.
Hi Hussaini,
Can i have a link to some of them
Is it possible Process Macro can be used for that
Will there be the need for justifying why Re or FE in running mediation with panel data?
I have used Process Macro for the mediation analysis and included year fixed effect (year dummies) and industry fixed effect (industry dummies). But i am struggling in justifying why fixed effect.
i dont know how to do Re and test using hausman.
Any help?
Yes I will send some papers to you. But please, can you explain the nature of your data it will help in selecting techniques and tools of analysis. You can easily conduct the pre estimation test to meet OLS assumptions (normality, multicolinearity, hetro and Hausman tests using Stata. These are possible in the sequence of regressions to meet the conditions of mediation in line with Baron and Kenny (1986) and Hayes et al. (2013).
@Lasisi Olalekan Isiaka, can you please explain the nature of your data?
Panel data by using Audit quality to mediate the relationship btw board characteristic and CEO pay or compensation
Thank you
I have done the mediation with process marco by Hayes in SPSS. My concern is justifying which panel model. FE or RE or Pooled.
If i test my individual relationship i can run panel (Fe, RE or pooled) and use the necessary tests to make the right choice (Hausman).
But in process macro there is nothing like FE, RE. is a pooled regression. Assuming my earlier test indicates i use RE. Then hoe w do i do that when running mediation. Or test says FE is better how do i run FE mediation. i will appreciate tour response
I will
OK, I understand, I think running fixed or random effects in mediation model can be done in the last sequence of mediation. That's when you run the dv and ivs while controlling for mediator. But actually I'm an expert in process macro. We wait for experts in spss to response.
@Lasisi Olalekan Isiaka. Use the following excel sheet as an example.
Hussaini Bala could you share me some of the papers that have mediation analysis with panel data, please
Ummi Junaidda Hashim , Hussaini Bala here is the example for cross-sectional data: https://data.library.virginia.edu/introduction-to-mediation-analysis/?fbclid=IwAR1xUaMSnTqBAXmrBCKDMTZnc41c7CaVJ-bF3MYMqXV7CWhSs4cqk46-KOk
It's in R. Package mediation works not just with cross-sectional data. You can perform mediation analysis, using package lme4, which is for multi-level data. You can elaborate regression models using lme4 and perform mediation analysis via function mediate.
You can see the difference between usual fixed and random effects models and multi-level data regressions here: https://stackoverflow.com/questions/49033016/plm-or-lme4-for-random-and-fixed-effects-model-on-panel-data
Hussaini Bala can I use the Process Macro in SPSS to test for mediation for my panel data? I have set up ids as you have shown in the demo file.
Sim, faça vários testes e crie instrumentos e metodologia. Criei a própria e inédita.
Hussaini Bala can you please send some papers using mediation with panel data (specifically, with fixed effects)? thanks!
Boa noite, é possível e tenho essa proposta na TESE DE DOUTORADO.
Article The mediating effect of financial performance on the relatio...
Aditi Kadam You can see this example:
https://stats.idre.ucla.edu/r/faq/how-can-i-perform-mediation-with-multilevel-data-method-2/
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running