For her dissertation research relating item content to slopes/loadings, my student is seeking papers that fit simple (i.e., no cross-loadings) structural models using IRT or CFA for Likert-style responses to measures of personality, attitude, etc. We will also fit such models ourselves if the data are made available to us.
In order to analyze the item content and relate it to the structural parameter, the items have to be provided or publicly available and the paper must provide the slopes/loadings.
We are particularly interested in measures of content that has shown wording effects like Rosenberg's self-esteem scale, big-five personality, positive and negative effect, etc. We cannot consider measures whose sole use is clinical (e.g., MMPI).
Hi Alan,
I have data for the Big Five that I would be happy to share with you. It's the same data that I presented at SIOP earlier this year. Let me know, I will email the files to you.
Check NHANES and NNHS (just google these terms, and you will find links to the CDC data repositories):
Both of these surveys are publicly available and have likert items related to physical functioning (how independent are you with walking?)
You can find a few studies doing factor analysis or IRT on the NHANES scale.
For the slightly bizarre, you can look at Likert scales that do not have a monotonic relationship with the underlying trait (called unfolding models). I think that these are conceptually fascinating but a bad idea practically speaking. Nevertheless, they have been statistically characterized with psychometric models:
http://www.psychology.gatech.edu/Unfolding/Intro.html
- Badges
- Science topic