Table 1
Item information statistics
Point of Maximum Information |
1. | - .5335 | 11. | - .3486 | 21. | -1.3415 |
2. | -1.8624 | 12. | - .2408 | 22. | .0778 |
3. | - .0595 | 13. | -1.1276 | 23. | - .8488 |
4. | - .1422 | 14. | -1.4116 | 24. | -1.1307 |
5. | - .3898 | 15. | -1.4472 | 25. | -1.3093 |
6. | -1.5271 | 16. | - .8430 | 26. | - .1914 |
7. | .0279 | 17. | - .6911 | 27. | - .6369 |
8. | -1.1277 | 18. | .1810 | 28. | - .6239 |
9. | -1.7692 | 19. | .2232 | 29. | - .7331 |
10. | - .2972 | 20. | .5614 | 30. | .3627 |
6.3.5 Factor analyses
Exploratory factor analysis. Exploratory principal-factor analysis was done, yielding 5
factors with eigenvalues greater than 1 in the Spanish sample (and .9 in the U.S. sample).
These factors accounted for roughly 50% of the variance in the Spanish data and 63%
of the variance in the U.S. data. Eigenvalues were 9.9, 1.5, 1.2, 1.2, and 1.1 respectively
for the Spanish data and 14.3, 1.5, 1.2, 1.0, and .9 in respectively for the U.S. data.
Confirmatory factor analysis. Confirmatory factor analysis also was done. The
conceptual framework proposed in Figure 1 provided a 3 x 5 model for the items of the
ESJI. We therefore evaluated corresponding three- and five-factor latent variable models
for the U.S. and Spanish data.
|