I carried out EFA and CFA to analyse the extent to which attitudes influence people's intention to ride a bicycle. Now in the SEM I want to include as mediator 'income' since my hypothesis is that social class (measured by the income group in which people are) influence their intention to cycle. To measure income, I have 5 different groups (based on the decils for income distribution). I carried out hyerarchical multiple regresion testing TPB and other additional constructs from other theories and in the final model I included income, but the contribution of income to explain the variance in intention was 1%. My sample size is 433 observations. Thank you very much
Thank you Heiko. My DV is intention (measured with 5 items) and the IV are attitudes (4 items), social image (3 items) and social comparison orientation (3 items).
So, in this mediation model, you are hypothesizing that attitudes, social image, and social comparison orientation are causes of income?
No, attitudes, social image and social comparison are the explanatory variables and intention is the response. For that, the model fit was achieved, but I want to check if income explains the relationship between the IV and the DV. Am I confusing the term mediator? Thank you
Dear Magda Capeda
try to check the data that I attach, you should enter another one mediating variables as shown below:
Hi Magda,
It seems that income may work as a moderator, since it may be strengthening or attenuating the relationship between each IV and the DV.
The mediation is more about an underlying explanatory process that elicits how the IV causes variation in the DV.
Hi,
My first thought was in line with what Luis just said: it looks like 'income' in Your case plays the role of the moderator as Your model may vary across selected income classes.
My guess is that is simply a matter of the classic mediator-moderator distinction issue. Thanks to everyone for pointing this out
Thank you very much for your responses Heiko, Rusiadi, Luis, Lukasz and Augustine. Reading more about the the difference between moderator and mediator (the Handbook on SEM from Zainudin Awang, have been very useful) I agree on having issues with the distinction. I will treat income as moderator and see the difference among the scioeconomic groups. I really appreciate your help, Kind regards
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis