ERIC Number: ED447150
Record Type: Non-Journal
Publication Date: 2000-Apr-26
Reference Count: N/A
Using Multilevel Structural Equation Modeling with Faculty Data.
Stapleton, Laura M.; Hancock, Gregory R.
This paper illustrates the differences in inference that can be seen when traditional and multilevel structural equation modeling techniques are applied to hierarchical data. Research on faculty is an area in which multilevel data exist, and where previous research generally has not modeled the nested structure. Using data from the National Study of Postsecondary Faculty (NSPOF), this paper demonstrates a method for analyzing data that contain measurement error and come from multilevel structures. The NSPOF database contains responses from 25,780 faculty randomly chosen from 817 participating institutions. The integration of multilevel regression modeling and structural equation modeling, which can facilitate proper inference, is used to study faculty satisfaction. In this study, no substantive differences were shown when traditional and more complex modeling techniques were compared, but the results from the analyses contribute to the knowledge base regarding the institutional and job characteristics that can affect faculty satisfaction. Two appendixes contain the NSPOF questionnaire and the EQS syntax for the "independence model." (Contains 2 tables, 7 figures, and 16 references.) (SLD)
Publication Type: Reports - Descriptive; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A