In a longitudinal study on the consequences of electronic health record implementation on healthcare professionals work motivation, we aim to test a first stage moderated mediation model, using GEE in SPSS. We want to follow the method described by Hayes (2015; index of moderated mediation). Although there is literature on using GEE for mediated effects (Schluchter, 2008), I was not able to find any (research) papers on using GEE for this specific purpose. Does anyone have experience with moderated mediation using GEE or know whether this is an appropriate method?
We have the following variables: (X = Time: before versus after EHR implementation), two mediators (M = job autonomy, task interdependence), one outcome variable (autonomous motivation) and one moderator (W = profession, with four subcategories). We anticipate that profession has a moderating effect on the relationship between time and the mediators.
I am afraid that this is not direct answer to your question but consider an alternative to GEE, - random effects modelling for longitudinal data -and why you may want to consider it
https://www.researchgate.net/post/Where_can_I_find_good_material_on_the_difference_between_mixed_models_and_gee_models
and then
Article Conceptualizing and Testing Random Indirect Effects and Mode...
and
Front. Psychol., 15 April 2020 | https://doi.org/10.3389/fpsyg.2020.00637
Article The Use of Mixed Models for the Analysis of Mediated Data wi...
You can use GEE.
BUT!
there is an important matter here, which is utilizing new tools and modern methods. I strongly recommend you follow Hayes's recommendations (2017/2019) with regard to mediation/moderation analysis. Additionally, using GEE is not really appropriate for your model. GEE is commonly used for parameter estimation for a linear model that may or may not include correlations among variables (as result). alternatively, if structure of covariance within your model is erroneous, then you may use GEE to solve some regularity issues.
I am positive that Hayes will fit your model much better than GEE.
Thanks for your suggestions! For those who are interested, so far I have discovered the following. It can be done using the GEE, when following the 'piecemeal' approach (which has the limitations that it is a bit outdated and increases chance of type 1 error).
The other method is to use Mplus, which Amanda Montoya (former PhD student of Hayes) describes in her thesis and for which she has developed a code, which can be found in her thesis (check her website).
Also, she has developed the MEMORE macro for SPSS, which can be used to test repeated measures mediation and moderation. So far, three models have been developed (unfortunately not the one I aim to test), but there are more models on the way.
Hi Jing Ren,
The name of the thesis is Conditional Process Analysis in Two-Instance Repeated-Measures Designs. You can find it here: https://etd.ohiolink.edu/apexprod/rws_olink/r/1501/10?clear=10&p10_accession_num=osu1530904232127584
good luck!
Gepke
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publication