WHich method of analysis

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Posts: 1
Joined: Fri Jun 03, 2016 5:27 am

WHich method of analysis

Postby Tyldo » Fri Jun 03, 2016 6:25 am

I have to make a hypothesis and run appropriate test on the follwing:

Imagine you are a 4th year psychology student studying factors affecting sexual health behaviours. You want to examine 3 variables in particular:

Sex (referring to male or female)
Condom usage, and
Fear of acquiring sexually transmitted diseases (STDs).
Your literature review suggests that sex (male or female) may predict condom usage, possibly because condoms are a male-oriented sexual health product. Further, you understand that the association between sex and condom usage may not be so simple, as it may be altered by other variables such as the individual’s fear of acquiring STDs.

I know the analysis can either only be ANCOVA, MANOVA, Correltion/Regression, or modertation/mediation analysis. However I am uncertain which. I am leaning toward multiple regression however, I want to keep it as simple as possible.

It is easiest to do a multiple regression, and report on the variance explained by both IV's, or considering I am measuring if there is a diff in condom usage between M and F, should I do something like an ANCOVA to determine if: DV- CONDOM USE differed between gender (IV - 2 levels, M &F) while controlling for covariate (Level of fear of STDS - measured by Likert scale, which is ordinal) However, given that the covariate is ordinal, this breaks ANCOVA assumption that covariate must be continuous.

Or is this a moderation analysis, of which we have only learnt this week, so seems unlikely given essay is due in 2 weeks, but university is brutal so not discounting it (still wrapping my head around how it works

Any guidance would be appreciated!


Who is online

Users browsing this forum: No registered users and 2 guests