Abstract
Some researchers suggest that panel data be used with simplex models in order to evaluate stability of opinions before drawing the conclusion that an attitude is detected. They have carried out studies showing that it is not always true that an attitude exists, mainly because the opinion is unstable. This paper proposes to continue this line of research, presenting both a new conceptualization of attitude stability and a way to evaluate it by using simplex models. The paper will show that when evaluating attitude stability, researchers' models must not only take into account the structure of the variances/covariances but also the structure of the means. The authors demonstrate that this new definition of change makes a difference in the conclusions with respect to the stability of opinions with respect to the role of women in society.
| Original language | English |
|---|---|
| Pages (from-to) | 269-285 |
| Number of pages | 17 |
| Journal | Quality and Quantity |
| Volume | 31 |
| Issue number | 3 |
| DOIs | |
| Publication status | Published - Aug 1997 |
Fingerprint
Dive into the research topics of 'Tests of stability in attitude research'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver