NotesFAQContact Us
Search Tips
ERIC Number: ED318756
Record Type: Non-Journal
Publication Date: 1990-Apr
Pages: 30
Abstractor: N/A
Reference Count: N/A
GENMOD, HLM, ML2 and VARCL, Four Statistical Packages for Hierarchical Linear Regression: Similarities and Differences.
Kreft, Ita G. G.; Kim, Kyung-Sung
A detailed comparison of four computer programs for analyzing hierarchical linear models is presented. The programs are: VARCL; HLM; ML2; and GENMOD. All are compiled, stand-alone, and specialized. All use maximum likelihood (ML) estimation for decomposition of the variance into different parts; and in all cases, computing the ML estimates involves complex non-linear expressions in the parameters. In all cases, the equations must be solved by an iterative procedure. The major differences among the programs lie in the choice of the criterion to optimize and the choice of the algorithm to optimize the criterion. For each program, the design philosophy, implementation details, and routines are specified. Use of these programs is illustrated with data from four sets: (1) the Second International Mathematics Study, using 3,691 cases; (2) data from 1,290 students in the Netherlands (the GALO study); (3) data from 5,310 secondary students in the Netherlands (the VOCAL study); and (4) data from 96 students from junior high schools in Los Angeles (California). These illustrations reveal the applicability of software for hierarchical linear regression. Ten tables contrast the results obtained by the various computer programs. (SLD)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: Netherlands Foundation for Educational Research, The Hague.
Authoring Institution: N/A