According to information provided by Dana Nunn, director of media relations at Colorado Mesa University, a study was performed to look at faculty gender equality in May of 2016. The study was performed by CMU’s Office of Institutional Research, Planning and Decision Support (IRPDS), and focused on evaluating the equity of salaries for CMU full-time faculty by gender.
The findings of the study show no inequity due to gender.
Two sets of salaries were analyzed in the study. The first was the 2015-16 salaries and the second was the preliminary 2016-17 salaries for all known faculty returners and new hires.
IRPDS ran an analysis of variance (ANOVA) test. ANOVA is used to determine statistical differences between the averages of two or more groups. This is done using a significance level, p-value and null hypothesis.
The significance level is typically set at 0.05 and a p-value is a number between zero and one that indicates the probability that something is likely or unlikely according to its value in relation to the significance level. Greater than the significance level means that something is likely and less than the significance level indicates that something is unlikely.
The null hypothesis often means the common view of something. When an ANOVA test is performed, a high p-value verifies that the null hypothesis is correct, or at least fails to refute the common view.
A p-value less than the significance level means a rejection of the null hypothesis. In this instance, the common view is proven to be incorrect.
IRPDS performed the ANOVA test using the null hypothesis that there existed no inequity to salary due to gender.
A factor included in the analysis was the rank of the faculty, which included instructor, assistant professor, associate professor and professor. Another factor was appointment type, meaning non-tenured, tenure-track or tenured.
The results of the statistical analysis showed that salaries across campus were impacted by the rank of the faculty member and the appointment type. Higher ranks and appointment types had higher salaries.
The p-values for gender, gender by appointment type and gender by rank were greater than the significant value for both sets of salaries. The lowest p-value regarding gender was gender by rank in the 2015-16 salary set at 0.151.
The p-value for gender without including rank or appointment type was 0.958 in the 2015-16 set and 0.717 in the 2016-17 set. IRPDS concluded the p-value for gender was “remarkably high in this analyses, thus not indicating an salary inequity due to gender.”
