Really didn't expect such a fast response, thanks!

The thing is (regarding hypotheses), my research is pretty much exploratory in nature, since I'm using two completely new instruments, but with much simpler research questions than the ones in the article. I'm not focusing on gender differences (even though I've included it as a control variable), but rather trying to explain the differences in attaining a leadership role with some of the old, and some of the new variables, which I'll describe, in short, to make sense of the correlation matrix in the attachment.

What I've taken from the attached article is the leadership role occupany (LRO) operationalization, meaning I want to:

(1) examine the role of all of these variables in predicting whether or not a person is in a leadership position (yes/no),

(2) and do the same thing, but for different levels, or hierachical categories of LRO (1-5, 5 being the highest).

Another similarity is the central variable, which is POWER MOTIVATION.

I've conducted an on-line survey and now have n=184 respondents' data on all of these variables:

- implicit power motivation, explicit power motivation, 6 personality traits (pretty much intuitive), social dominance orientation, environmentalism, and some demographic variables.

I'm planning on using environmentalism as another criteria variable, besides leadership role occupancy, but I'll get to that later, so just ignore it for now

I have a basic understanding of the logit being the natural log of the odds ratio, and how it relates to the independent variables, or predictors, regarding ONE predictor and one (binary) outcome, but with this much possible predictors, and especially a 5-degree outcome, I'm totally lost. This is mostly because I'm not familiar with MLE, and there are no standardised beta-coefficients for me to compare, like in linear regression, so conducting a logistic regression using binary LRO as a dependent variable, and either combination of the rest of the variables as covariates, doesn't really help me understand anything in terms of relationships, or effect sizes, except for odds ratios for those variables, keeping the others constant.

What I'm really asking here, before I confuse you guys more than myself, is how do I look at my data prior to during conducting LRA in order to figure out if there are any patterns or effects such as supression, which could affect the interpretation? And are there variables I should not even include in the analysis beyond bivariate correlation?

And to think my professor thinks I'm perfectly capable of answering all of the above myself