This report explains why the Management Implications of Team Teaching (MITT) project chose multiple linear regression and path analysis to analyze through-time relationships among variables, and why it rejected repeated-measures analysis of variance (ANOVA) and difference scores over time. Project MITT examined governance and work structures for five time periods from 1974 to 1976 in 29 elementary schools, 16 of which had introduced team-teaching (or unitized) methods in 1974. To analyze longitudinal changes among variables and schools, the project's statistical techniques had to take account of small sample size and multiple time periods; they also had to control for pre-1974 differences among the schools, changes in variables because of unitization, and differences in variable means and ranges. All of these factors interfered with comparisons of unitized and nonunitized schools and distorted relationships among the variables. Hierarchical multiple linear regression solved these problems by relating variables to one another both over time and in order of explanatory power. Path analysis using lagged multiple linear regression helped to test postulated relationships through time and explore for further relationships. Four appendices discuss ANOVA, difference scores, path analysis, and corrections used for data cyclicity. (Author/RW)