The purpose of this study was to determine whether selected environmental variables could be used to differentiate public two-year colleges on the basis of collective-bargaining status and bargaining-agent affiliation. The following null hypotheses were tested:

1. There are no differences between states with collective bargaining in community colleges and those without collective bargaining, as measured by the set of environmental variables selected for this study.

2. There are no differences between states with enabling legislation for collective bargaining in community colleges and those without legislation, as measured by the set of environmental variables selected for this study.

3. There are no differences between community colleges which are unionized and those which are not unionized, as measured by the set of environmental variables selected for this study.

4. There are no differences among community colleges affiliated with, respectively, the American Federation of Teachers, the National Education Association, and with no union, as measured by the set of environmental variables selected for this study.

Hypotheses Three and Four represented the principal analyses.

A population of 556 colleges, 239 of which were unionized, was used. Data were collected on 29 environmental variables and analyzed by stepwise discriminant analysis procedures from the Statistical Package for the Social Sciences. Each principal analysis was performed twice: once with nonunion colleges selected from all states, and once with nonunion colleges restricted to unionized states. Each null hypothesis was rejected at a level of significance of 0.05 of less. Although each analysis produced its own linear combination of discriminating variables, seven variables were common to all principal analyses. The environmental influences represented by these variables were: (a) the level of control over the college, (b) whether the college is part of a system, (c) whether the state has enabling legislation, (d) the proportion of the reference area population living in urban areas, (e) value added by manufacture in the reference area, (f) real per capita income in the state, and (g) union membership as a percent of the state's nonagricultural labor force.