NotesFAQContact Us
Search Tips
ERIC Number: ED453261
Record Type: Non-Journal
Publication Date: 2001-Apr-11
Pages: 24
Abstractor: N/A
The Effects of Centering Method on Parameter Estimates and Variance Explained in Multilevel Models.
Koenig, Judith A.; Lissitz, Robert W.
This paper uses data from the National Education Longitudinal Study of 1988 (NELS:88) dataset to demonstrate practical examples of the ways in which the method used for centering level-1 variables in multilevel models affects the findings. Demonstrations compare raw metric scaling, grand mean centering, and group mean centering for successively more complex models. Comparisons are made of parameter estimates, their significance levels, and increments in variance explained. Findings show that results are generally similar for raw metric scaling and grand mean centering, and these results differ from those obtained under group mean centering. Two methods are demonstrated for estimating incremental variance explained by nested models. The ways in which centering can be used to examine between-groups and within-groups effects are also shown. (Contains 6 tables, 2 figures, and 11 references.) (Author)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A