0
$\begingroup$

I'm having some issues running an analysis with two categorical variables that are both non-dichotomous...

Some background information on my study: My study focuses on how young adults sexually communicate in an online dating setting, and what influences them to do so. Participants are given a simulated online dating scenario where they see an online dater's profile (with a picture and a message), and then they are given the opportunity to reply to the profile.

What I'm having trouble with is analysing how the profile's message (sorted into 4 levels based on how 'sexy' it is) correlates to how the participants responds (also coded into 4 levels of how 'sexy' their response is). I hypothesise that the 'sexier' the profile's message is, the more 'sexy' the participant will respond. ​

So my two categorical variables are: How sexy the profile's message is (4 levels) How sexy the participants response is (4 levels)

What analysis should I run for this??

$\endgroup$
2
$\begingroup$

I think you want the Spearman correlation. This page has a discussion as well as how to perform the analysis in SPSS:

https://statistics.laerd.com/spss-tutorials/spearmans-rank-order-correlation-using-spss-statistics.php

$\endgroup$
0
$\begingroup$

If the 4 categories of each scale are numbers where the distance between each value is assumed to be equal, then you have interval level data and you can try a pearson regression although truncating the scale to 4 levels might create problems. If the 4 categories are not numeric but were assigned verbal descriptions then you probably have an ordinal scale in which case spearman ranks will be appropriate. A final option is to treat the data as nominal and run a chi-square test for independence (called crosstabs in SPSS).

$\endgroup$
0
$\begingroup$

Davester is right, but better go with his second option 'Chi Square'. Most likely you will eventually have those variables measured on a ordinal level (instead of interval)..

$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.