I am looking to investigate the influence of personality traits at the item level on a behaviour which has only one categorical level. Every subject in the sample will have displayed this type of behaviour, so the study is to examine which of the traits are predictive of this behaviour.
This would mean that both the independent and dependent variables would be categorical, it looks as if regression would have been a candidate however I am not aware of any regression models which the outcome variable can be a single level categorical variable?
Any help would be useful!
Sorry but no hope. you need at least 2 levels for your dependent variable. In that case logistic regression would be the method. best,David Booth BTW aren't there some individuals that don't have this trait somewhere?
A simple polychoric correlation (pairwise association) between all traits and the single DV variable will give you an indication of the associative strength.
Hope this helps,
Matt
Thanks for the responses guys that was a very fast response. As there are actually a number of categories of the behaviour in question it looks as it I could try either regression or anova on this data?
David, I do not believe there will be any individuals who would score zero on the items due to the close match between the behaviour of interest and the measures being used.
That Matthew, I was looking for an overall measure of association and that sounds like the way to go.
Thanks to you both.
If you can find the additional information I would recommend logistic regression. See the attachment. Best regards, David
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling