Problems of analysis are discussed in the three areas of estimating pupil gain, grouping observation items into measures, and relating gain to classroom behavior. Procedures deemed most hopeful are that gain estimates should be based on at least two covariates; that observation items should be aggregated using factor analysis, a combination of logical and factor analysis, or clustering or factoring teachers; that process-product relations should be analyzed by multiple regression, with at least a two-stage model that uses both classrooms and individual pupils as the unit of analysis.
Get full access to this article
View all access options for this article.
References
1.
BennettN.Teaching styles and pupil progress.Cambridge: Harvard University Press, 1976.
2.
BereiterC.Some persisting dilemmas in the measurement of change. In HarrisC. W. (Ed.), Problems in measuring change.Madison, Wis.: University of Wisconsin Press, 1963.
3.
CampbellD. T., & ErlebacherA.How regression artifacts can mistakenly make compensatory education look harmful. In HelmuthJ. (Ed.), Disadvantaged child (Vol. 3), pp. 185–210. New York: Brunner/Mazel, 1970.
4.
CohenJ.Multiple regression as a general data-analytic system. Psychological Bulletin, 1968, 70(6), 426–443.
5.
CohenJ., & CohenP.Applied multiple regression/correlation analysis for the behavioral sciences.New York: John Wiley & Sons, 1975.
6.
CokerH.An effectiveness-directed approach to teacher evaluation and certification. Paper presented at American Association of Colleges for Teacher Education, Chicago, February 1978.
7.
CronbachL. J.Research on classrooms and schools: formulation of questions, design, and analysis.Stanford, Calif.: Stanford Evaluation Consortium, Stanford University, 1976. (ERIC Document Reproduction Service No. ED 135 801)
McLeanJ. E.An empirical examination of analysis of covariance with and without Porter's adjustment for a fallible covariate. Unpublished doctoral dissertation, University of Florida, 1974.
18.
McNemarQ.On so-called test bias. American Psychologist, 1975, 30, 848–851.
19.
MedleyD. M., & MitzelH. E.Some behavioral correlates of teacher effectiveness. Journal of Educational Psychology, 1959, 50, 239–246.
20.
MorrisJ. D.A comparison of regression prediction with data-level variables versus factor scores. Unpublished doctoral dissertation. University of Florida, 1975.
21.
PeckhamP. D., GlassG. V., & HopkinsK. D.The experimental unit in statistical analysis. Journal of Special Education, 1969, 3, 337–349.
22.
RathsJ.The appropriate experimental unit. Educational Leadership, 1967, 25, 263–266.
23.
SoarR. S.An integrative approach to classroom learning. (NIMH project numbers 5-R11 MH 01096 and 7-R11 MH 02045.) Philadelphia: Temple University, 1966. (ERIC Document Reproduction Service No. ED 033 749)
24.
SoarR. S.Follow Through classroom process measurement and pupil growth (1970–71): final report.Gainesville, Fla.: Institute for the Development of Human Resources, University of Florida, 1973. (ERIC Document Reproduction Service No. ED 106 297)
25.
SoarR. S., & SoarR. M.An empirical analysis of selected Follow Through programs: an example of a process approach to evaluation. In GordonI. J. (Ed.), Early childhood education.Chicago: National Society for the Study of Education, 1972.
26.
SoarR. S., & SoarR. M.Classroom behavior, pupil characteristics, and pupil growth for the school year and for the summer.Gainesville, Fla.: Institute for the Development of Human Resources, University of Florida, 1973. JSAS Catalog of Selected Documents in Psychology 5 (1975): 200. (MS. No. 873)
27.
SoarR. S., & SoarR. M.Emotional climate and teacher management: a paradigm and some results. In WalbergH. J., & PetersonP. L. (Eds.), Conceptions of teaching.National Society for the' Study of Education and McCutchan Publishing Corp., in press.
28.
WalbergH. J.Generalized regression models in educational research. American Educational Research Journal, 1971, 8, 71–91.
29.
WileyD. E., & BockR. D.Quasi-experimentation in educational settings: comment. The School Review, Winter 1967.
