NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED401329
Record Type: Non-Journal
Publication Date: 1995-Jun-23
Pages: 36
Abstractor: N/A
Reference Count: N/A
ISBN: N/A
ISSN: N/A
Is More Ever Too Much: The Number of Indicators per Factor in Confirmatory Factor Analysis.
Marsh, Herbert A.; And Others
Whether "more is ever too much" for the number of indicators (p) per factor (p/f) in confirmatory factor analysis (CFA) was studied by varying sample size (N) from 50 to 1,000 and p/f from 2 to 12 items per factor in 30,000 Monte Carlo simulations. For all sample sizes, solution behavior steadily improved (more proper solutions and more accurate parameter estimates) with increasing p/f. There was a compensatory relation between N and p/f; large p/f compensated for small N and large N compensated for small p/f, but large N and large p/f was best. A bias in the behavior of the chi square was also demonstrated where apparent fit declined with increasing p/f rations even though the models were all "true." Fit was similar for proper and improper solutions, as were parameter estimates from improper solutions not involving offending estimates. The 12-p/f data were also used to construct 2, 3, 4, or 6 parcels of items (e.g., 2 parcels of 6 items per factor, 3 parcels of 4 items per factor, etc.), but the 12-indicator (nonparceled) solutions were somewhat better behaved. The study shows that traditional "rules" implying fewer indicators should be used for smaller N may be inappropriate and that CFA researchers should use more indicators per factor than is evident in current practice. (Contains 4 figures, 5 tables, and 41 references.) (Author/SLD)
Publication Type: Reports - Evaluative
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A