The Impact of Accountability: An Evaluation of Performance-Based Budgeting in the North Carolina Community College System

Abstract

The purpose of this study was two-fold: first, to review and determine the rate of progress, if any, among North Carolina community colleges in meeting the five core performance measures; and, second, to determine if a variety of independent variables have the potential to impact a community college's performance with respect to these mandated standards.

Descriptive statistics were used first to analyze the rate of progress among North Carolina community colleges on each of the five core performance measures. A review of the results from 1998-2001 demonstrated little variance in performance.

Additional analysis was conducted using both bivariate and multivariate regression. Pearson Correlation Coefficient was used to determine if a relationship exists between each of the core performance measures and a variety of independent variables identified as student and institutional characteristics. At an alpha level of .05 significance, several independent measures demonstrated notable significant relationship with two of the performance measures:

• Performance measure 2: Ten of the 14 independent variables demonstrated primarily moderate, but significant correlations with the dependent variable or the passing rate for first-time test-takers on licensure and certification examinations; and

• Performance measure 5: Two of the 14 independent values demonstrated moderate, but significant correlations.

The last test administered was a multiple regression analysis, which examined the interaction between each performance measure and the independent variables for indications of sample variance. The model was able to explain sample variance for two of the five core performance measures. For the second performance measure, the stepwise regression model suggests that almost half of the sample variation (49%) can be explained by four of the independent variables. Approximately a third (31%) of the sample variance for the fifth performance measure was explained by three of the independent variables.