Mediation Effect using Panel data

[Farheen Akram]

Hi
i am going to use mediation (independent, mediator and dependent variables are continuous). but i read various types of mediation in Stata. i am confused which one is the best suitable for my data and model. I am interested in direct and indirect paths.

[Roman Mostazir]

For assessing the direct and indirect effect simultaneously, use Stata's sem or gsem (for multi-level data) command which stands for Structural/Simultaneous Equation Model. Type help gsem. You would be better off with sem first as it gradually will build up your knowledge. It will take you through how to test the non-linear indirect effects. Alternatively, there is a user written programme called ml_mediation by Krull & MacKinnon (2001) and can be used with continuous mediator. Type findit ml_mediation in your Stata command box. There might be other user written programmes for multi-level data but I am no aware of.

[Farheen Akram]

Thank you so much for your information.

[Amedeo Pugliese]

Hi Roman,
this is perhaps not common practice - not allowed in the FORUM - but I thought you might have some ideas about the following, related, topic that I posted a month ago on STATALIST but no replies.
Hi Statalisters,

I am trying to fit what I thought would be a simple multilevel mediation model of the type 2-2-1 (Preacher et al, 2010; Krull & MacKinnon, 2001 and many others).

This is what the variables look like:
DV is a level 1
IV is a level 2
MV is a level 2
CNT1 is a level 1
CNT2 is a level 2
org_id is the level 2 clustering variable

For the sake of completeness, I have data on directors nested within 65 boards. Data are taken at the same point in time and are all observed data (e.g. no latent).

I have some questions that are bugging me as I do not have access to MPLUS that seems used largely in prior studies fitting 2-2-1 models (Super et al 2016; Lehman-Willenbrok et al 2015).

I guess the main question is: which command should I use in STATA 14.2? I have tried with the following ones but not sure which one is more appropriate in the estimation to avoid conflating results:

1. ml_mediation

code:
ml_mediation, dv(dv) iv(iv) mv(mv) l2id(org_id) cv(CNT1 CNT2)

When I try to estimate the model I do not succeed (e.g. Iteration 59: log likelihood = 8131.7637 (not concave)).
It seems to me that given the path a is a 2-2 (MV on IV) the model cannot converge.

Next I try to bootstrap as follows:
bootstrap indeff=r(ind_eff) direff=r(dir_eff) toteff=r(tot_eff), ///
reps(500) seed(1) cluster(org_id) idcluster(norg_id): ///
ml_mediation, dv(dv) iv(iv) mv(mv) l2id(org_id) cv(CNT1 CNT2)

estat boot, percentile bc

2. xtmixed

I get the same when I try to use xtmixed and specify the clustering at the org_id level. The problem occurs in the estimation of path a (MV --> IV) whereas I am able to estimate Eq2 and Eq3 using the xtmixed, rmle.

The code is as follows:
*Eq - path c (total effect) (DV --> IV)
mixed dv iv CNT1 CNT2 || org_id:, reml

*Eq - path a (MED --> IV)
mixed MV IV CNT1 CNT2 || org_id:, reml

** this one cannot be estimated. Only if I remove the clustering it will work. **

*Eq - path b (DV --> MED) & c' (DV --> IV)
mixed DV MV IV CNT1 CNT2 || org_id:, reml

3. GSEM

I have tried with the GSEM routine

gsem (MV <- IV CNT1 CNT2) (DV <- MV IV CNT1 CNT2 M1[org_id])

I can estimate this one but if I introduce clustering in the first bracket - nothing works.
Then I test the indirect/direct effects as follows:
nlcom _b[DV:MV]*_b[MV:IV]

nlcom _b[DV:IV] + _b[DV:MV]*_b[MV:IV]

Could you please help me understand which one of these is correct and what I am doing wrong?

The alternative I see is using a MLM with estimation of two models as separate (Chen et al 2007 J of App Psy is a great example but the execution seems old style).

Best regards and I am extremely thankful for your support.
Amedeo

[Roman Mostazir]

Amedeo, I am afraid I need to say that you should have started a new thread rather killing the thread posted by Farheen. Answering your question will not answer the question asked by the original poster whereas the poster spent time on posting this and we all need to respect that. I replied to your original post. Have a look.
best,

[Amedeo Pugliese]

My apologies Roman. I had a thread but no replies and thought this was on a similar topic.
Apologies again

[julian mwanana]

Greetings,
Can anyone share with me a good sample journal paper on how to present mediation results? i mean how regression table with mediation should look like( for such simple model like this x-m-y)