review of causal mediation software, including Stata

[Rich Goldstein]

for those who might be interested, the current issue of Structural Equation Modeling: A Multidisciplinary Journal, Volume 27, Issue 6, November-December 2020

includes a comparative review of causal mediation software: Causal Mediation Programs in R, Mplus, SAS, SPSS, and Stata
Matthew J. Valente , Judith J.M. Rijnhart , Heather L. Smyth , Felix B. Muniz & David P. MacKinnon
Pages: 975-984

my library does not have access to this journal so I have not read the artaicle

[ericmelse]

Dear Rich,

You have access here.

Best,
Eric

[Rich Goldstein]

umm... thank you but this appears to be either paying for the article (not interested) or something I have tried and was told was not available to me (or some such language)

let me clarify that my comment in #1 was NOT meant to suggest that I should somehow get access to the article; it was meant as an explanation of why I was not mentioning what the article had to say about Stata

[Roman Mostazir]

why I was not mentioning what the article had to say about Stata

It really is not a comparison of software rather user written programmes. Stata is discussed here in reference to 'paramed' programme written by Richard Emsley and authors seem totally unaware that Stata's own official 'sem' suit can be used for 'mediation analysis'.

[Rich Goldstein]

I have been sent the article by other Stata users - thank you

just to add to Roman Mostazir s comment - the article also briefly mentions a second user-written program (med4way); I agree that it is unfortunate that there was no discussion of -sem-

[Tany Garciduenas]

Hello.
I've started using the command "med4way" recently for mediation analyses. The issue is that when I indicate the categorical covariables with the prefix i. , STATA doesn't run the syntaxis.
Here's an example:
med4way sdq_total_00 ST_tot_00 sueno_00 i.ajuste_centro edad_vs i.sexo_vs zIMC_00 imc_madre imc_padre i.educa_m_00 i.educa_p_00 i.recom_lacta_materna_SN ///
ffq_energia_00 act_fisica i.edo_ocup_madre i.edo_ocup_padre, yreg(linear) mreg(linear) a0(2) a1(0) m(10.38453)

The error that results from this is as follows:
factor-variable and time-series operators not allowed
r(101);

So, I wonder if is neccessary to indicate the categorical confounder variables with this prefix when using the command "med4way"..
Could please help me with this?

[Carlo Lazzaro]

Tany:
welcome to this forum.
Please start a new thread. Thanks.