Dissertation Analysis Confusion

#1
Hello all and thanks for any future responses.

I am doing a study looking at whether three independent categorical/dichotomous variables (gender, Greek status, and class status (i.e. freshman vs senior)) influence a person's disordered eating behaviors and attitudes among college students (continuous using EDE-Q). This was simple. However, I decided to include a few other measures (STAI for anxiety/stress, PSPS for perfectionism, and OBCS for body image) which are continuous, to address factors that commonly influence/exacerbate eating disorder symptoms. I was wondering if a multivariate multiple regression or a factor analysis would be better to use. My hypotheses include: females will have more disordered eating (DE) than males, freshmen will have more DE than females, and students in greek life will have more DE than traditional students. So the main variable I am wanting to measure is disordered eating (EDE-Q) but I want to have the stress, perfectionism, and body image data to better explain the differences between the groups.

What do you all think? Thanks!
 

Karabiner

TS Contributor
#2
I want to have the stress, perfectionism, and body image data to better explain the differences between the groups.
Do you mean that you wnat to perfor mediator analyses, such as female gender -> more perfectionism -> more disordered eating attitudes ?
This should be possible, if sample size is large enough.

With kind regards

Karabiner