Funding models of community colleges in 10 Midwest states

Community College Review, Winter, 2004 by Carol Piper Kenton, John H. Schuh, Mary E. Huba, Mack C. Shelley, II

The states were grouped by their relative dependency on each of these three revenue sources using the scheme depicted in Table 2. The levels of dependency and their associated percentages are as follows: Very High (40% or more of their revenue came from this source), High (dependency between 30% and 39%), Moderate (20% to 29%), Low (13% to 19%), Very Low (3% to 12%), and Extremely Low (0% to 2%). When at least two of the three states fell in the same category, that category was selected as the level of dependency.

For each model, averages were calculated for each of the three funding sources, and the data are shown in Table 3. As can be seen, the averages fall within the ranges set for each dependency category described above.

For each of the three sources of current funds revenue that were used in defining the models, a data set containing the mean percentages for the states within each model was created. For each revenue source, a one-way ANOVA was used to determine if the four models depicted in Table 3 were significantly different from one another. Results of these tests are shown in Table 4. The significant F-ratios indicate that at least one pair of models differs significantly for each revenue source.

Tukey post hoc test results are shown in Table 5. The test for state appropriations indicated that there were significant differences between Model 1 and all other Models (p < .01). Models 2 and 3 were significantly different from each other (p < .01), but not significantly different from Model 4 (p = .09 and p = . 11, respectively). In the tuition and tees category, Models 1, 2, and 3 were significantly different from each other (p < .01 top = .02) but Model 4 was not significantly different from any other Model (p = .06 to p = .98). The Tukey follow-up tests for local appropriations indicated that there was a significant difference between all pairings of Models (p < .01 top = .01) except Model 1 and Model 2 (p = .08).

The second research question examined how various funding models have provided sustained or increased revenue over the decade of the 1990s. For each model shown in Table 2, the average amount of total current funds revenue dollars per institution in the states represented by the model was calculated for 1990 and 2000, and the percentage change between the 2 years was calculated, as seen in Table 6. For example, Model 1 includes the states of Minnesota (18 institutions included in the study), Indiana (14 institutions), and Ohio (41 institutions).

In Minnesota mean funding per institution was $12,062,330 (total current funds of $217,121,948 in 1990 divided by 18 institutions). In Indiana the mean per institution was $11,541,232 (total current funds of $161,577,251 divided by 14 institutions), and in Ohio the mean per institution was $12,817,664 (total current funds of $525,524,219 divided by 41 institutions). The total by combining the means from each of the states was $36,421,226 for the three states included in the model for 1990. Similar calculations were made for the other states in 1990, and then the total was calculated in 2000 to reflect the growth over the decade. This total then was compared to the change in the Higher Education Price Index (HEPI) between 1990 and 2000 (39.84%). The right-hand column of Table 6 shows that all of the funding models generated considerable revenue in excess of the change in the HEPI. Model 1 (very high state appropriations, moderate tuition and fees, and extremely low local appropriations) generated the greatest increase in current funds revenue, whereas the balanced approach used only in Michigan ("moderate" reliance on all three sources) generated the smallest increase in current funds revenue.