I am struggling to interpret the results of a binomial logistic regression I did.
The experiment has 4 conditions, in each condition all participants receive different version of treatment.
DVs (1 per condition)=DE01,DE02,DE03,DE04,
all binary (1 - participants take a spec. decision, 0 - don't)
Predictors: FTFinal (continuous, a freedom threat scale)
SRFinal (continuous, situational reactance scale)
TRFinal (continuous, trait reactance scale)
SVO_Type(binary, egoists=1, altruists=0)
After running these binomial (logit) models,

model_soc_inf<- glm(mydata$DE01~FTFinal+SRFinal+TRFinal+SVO_Type, family=binomial(link='logit'),data=mydata) model_soc_inf1<- glm(mydata$DE01~FTFinal+SRFinal+TRFinal, family=binomial(link='logit'),data=mydata) summary(model_soc_inf) model_pers_inf <- glm(mydata$DE02~FTFinal+SRFinal+TRFinal+SVO_Type, family=binomial(link='logit'),data=mydata) model_pers_inf1 <- glm(mydata$DE02~FTFinal+SRFinal+TRFinal, family=binomial(link='logit'),data=mydata) model_pers_inf2 <- glm(mydata$DE02~SRFinal+TRFinal+SVO_Type, family=binomial(link='logit'),data=mydata) summary(model_pers_inf) model_soc_uninf<-glm(mydata$DE03~FTFinal+SRFinal+TRFinal+SVO_Type, family=binomial(link='logit'),data=mydata) model_soc_uninf1<-glm(mydata$DE03~FTFinal+SRFinal+TRFinal, family=binomial(link='logit'),data=mydata) summary(model_soc_uninf) model_pers_uninf<-glm(mydata$DE04~FTFinal+SRFinal+TRFinal+SVO_Type, family=binomial(link='logit'),data=mydata) model_pers_uninf1<-glm(mydata$DE04~FTFinal+SRFinal+TRFinal, family=binomial(link='logit'),data=mydata) summary(model_pers_uninf)

I ended up with the followingenter image description here. Initially I tested 2 models per condition, when condition 2 (DE02 as a DV) got my attention. In model(3)There are two variables, which are significant predictors of DE02 (taking a decision or not) - FTFinal and SVO Type. In context, the values for model (3) would mean that all else equal, being an Egoist (SVO_Type 1) decreases the (log)likelihood of taking a decision in comparison to being an altruist. Also, higher scores on FTFinal (freedom threat) increase the likelihood of taking the decision. So far so good. Removing SVO_Type from the regression (model 4) made the FTFinal coefficient non-significant. Removing FTFinal from the model does not change the significance of SVO_Type.

So I figured: ok, mediation, perhaps, or moderation. I tried first to look for mediation in both in R and SPSS. The moderation attempt was in vain: entering an interaction term SVO_Type:FTFinal makes all variables in model(3) non-significant. Here's the code for that:model1<-glm(DE02~FTFinal,family=binomial(link='logit'),data=mydata) summary(model1) model2<-glm(DE02~SVO_Type,family=binomial(link='logit'),data=mydata) summary(model2) model3<-glm(DE02~FTFinal+SVO_Type,family=binomial(link='logit'),data=mydata) summary(model3) interaction<-glm(DE02~SVO_Type+FTFinal+SVO_Type:FTFinal, family =binomial( link = "logit"),data = mydata)
As for mediation, I followed this mediation procedure for logistic regression, but found no mediation.

To sum up: There is some relationship between SVO_Type and FTFinal, but I have no clue what. Predicting DE02 from SVO_Type only is not significant. Predicting DE02 from FTFinal is not significant Putting those two in the regression makes them both significant predictors. Including an interaction between these both in any model, predicting DE02 model makes all variables in the model insignificant.
So I am at a total loss: As far as I know, to test moderation, you need an interaction term. This term is between a categorical var (SVO_Type) and the continuous one (FTFinal), perhaps that goes wrong? And to test mediation outside SPSS, I tried the "mediate" package in R, only to discover that there is a "treatment" argument in the main function, which is to be the treatment variable (exp Vs cntrl). I don't have such, all ppns are subjected to different versions of the same treatment. I apologize for this external way of uploading the dataset, it is way too complicated to reproduce here (I am a noob). Any help would be greatly appreciated. I have no clue what the relationship between SVO_Final and FTFinal is.

  • $\begingroup$ It would be helpful to pare down all unnecessary information in this post, like all the DVs you're not asking about. Regarding mediation, the "treatment" variable just means the predictor (i.e., "X") in the mediation graph. It doesn't have to do with the design of your study, where "treatment" might take a different meaning. $\endgroup$ – Noah Jul 25 '18 at 18:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.