Faculty Pay Equity Study - Morgan State University



center228600Faculty Pay Equity StudyPrepared for Morgan State UniversityPatrina Clark, President, Pivotal Practices Consulting, LLCAugust 1, 2014Project Number 14/PRO-2008-S (E-Maryland Solicitation MDR 1331011071)center7772400Table of Contents TOC \o "1-3" \h \z \u Executive Summary PAGEREF _Toc394927945 \h 1Introduction PAGEREF _Toc394927946 \h 4Background PAGEREF _Toc394927947 \h 4Study Methodology PAGEREF _Toc394927948 \h 5Policy Analysis PAGEREF _Toc394927949 \h 6Internal Policies and Practices Survey PAGEREF _Toc394927950 \h 6Internal Policies and Practices Documentation Review PAGEREF _Toc394927951 \h 8Summary of Results PAGEREF _Toc394927952 \h 9Statistical Analysis PAGEREF _Toc394927953 \h 9Comparison of MSU Faculty Gender Counts and Salaries to National Averages PAGEREF _Toc394927954 \h 9Comparison of Salary Averages PAGEREF _Toc394927955 \h 9Composition of the Faculty Pool PAGEREF _Toc394927956 \h 11Data Analysis and Methodology PAGEREF _Toc394927957 \h 12Assumptions and Data Limitations PAGEREF _Toc394927958 \h 13Demographic Data Analysis PAGEREF _Toc394927959 \h 13Univariate, Bivariate, and Multivariate Analyses PAGEREF _Toc394927960 \h 14Tenure Status PAGEREF _Toc394927961 \h 14Average Years of Service PAGEREF _Toc394927962 \h 15Tenure and Average Years of Service PAGEREF _Toc394927963 \h 15Faculty by College (Discipline) PAGEREF _Toc394927964 \h 15Regression Analysis - Overview PAGEREF _Toc394927965 \h 17Regression Analysis – Results PAGEREF _Toc394927966 \h 18Transactional Data Analysis PAGEREF _Toc394927967 \h 22Summary of Results PAGEREF _Toc394927968 \h 24Recommendations PAGEREF _Toc394927969 \h 26Salary Compression PAGEREF _Toc394927970 \h 27Appendix A: References PAGEREF _Toc394927971 \h 1Appendix B: Faculty and Staff Recommendations for Process Improvement PAGEREF _Toc394927972 \h 1Appendix C: Regression Analysis Summary Output Tables PAGEREF _Toc394927973 \h 1GENDER PAGEREF _Toc394927974 \h 1GENDER AND TENURE PAGEREF _Toc394927975 \h 3GENDER, TENURE, YEARS OF SERVICE PAGEREF _Toc394927976 \h 5GENDER, TENURE, YEARS OF SERVICE, SCHOOL PAGEREF _Toc394927977 \h 7List of Tables:Table 1: Internal Policies and Practices Survey Summary7 TOC \h \z \c "Table" Table 2: Education Level Doctorate, National Public Institutions Comparison9Table 3: Education Level Master’s, National Public Institutions Comparison10Table 4: Education Level Doctorate, Regional Public Institution Salary Comparison10Table 5: Education Level Master’s, Regional Public Institution Salary Comparison10Table 6: Average Salaries, Standard Deviation, and Expected Salaries - By Rank13Table 7: Salaries, by Rank and Gender, Outside Expected Range14Table 8: Number and Percentage of Tenured and Non-Tenured Faculty, by Gender15Table 9: Average Years of Service, by Rank and Gender PAGEREF _Toc389054801 \h 15Table 10: Tenure and Average Years of Service, by Rank and Gender PAGEREF _Toc389054802 \h 15Table 11: MSU Faculty by College, Rank, and Gender16Table 12: Average Annual Salary by College (Discipline), Rank, and Gender17Table 13: MSU Enrollment by School/College, 200617Table 14: Regression Results - Gender PAGEREF _Toc389054806 \h 19Table 15: Regression Results - Gender and Tenure20Table 16: Regression Results - Gender, Tenure, and Years of Service21Table 17: Regression Results - Gender, Tenure, Years of Service, and School22Table 18: Number and Average Salaries of New Hires, by Rank and Gender PAGEREF _Toc389054810 \h 23Table 19: Number of Promotions by Gender and Effective Date (Year) PAGEREF _Toc389054811 \h 23Table 20: Adjustments by Type, Number, and Average % Increase PAGEREF _Toc389054812 \h 24List of Figures: TOC \h \z \c "Figure" Figure 1: Post-Secondary Faculty by Gender, National Totals PAGEREF _Toc389055067 \h 11Figure 2: Faculty by Gender (2014), MSU PAGEREF _Toc389055068 \h 12Executive SummaryIntroduction Morgan State University (MSU) contracted with Pivotal Practices Consulting LLC (PPC), a Maryland-based management consulting firm, to conduct a faculty pay equity study for its tenured and tenure-track (Assistant Professor, Associate Professor, and Professor) faculty to:Determine whether there are gender-based salary inequities.Make recommendations to resolve any such inequities.Methodology and AnalysisPPC developed a study project plan consistent with widely accepted practices for conducting academic pay studies for higher education faculty. PPC conducted quantitative and qualitative data analysis that included the following elements:Policy and Procedures ReviewInternal SurveyStatistical AnalysisThe study’s methodology was applied through the following analyses:Internal Policies and Practices SurveyInternal Policies and Practices Documentation ReviewGeneral Statistical AnalysisRegression AnalysisKey FindingsThe following are some of the key findings of the study:There is no evidence of gender-based salary inequities. Further, PPC found no meaningful relationships between tenure, years of service, and salary.While beyond the scope of the study, PPC did find disparities in salaries attributable to factors other than gender among MSU faculty, most notably School/College affiliation and academic discipline.In comparing MSU faculty salaries with salary averages at the national level:Average salaries for MSU female faculty at the Assistant Professor level were significantly higher than the national average.The salary gaps between male and female faculty at the Associate Professor and Professor levels were smaller for MSU faculty than for faculty at the national level.In comparing MSU faculty salaries with average salaries at the regional level:For faculty with a master’s degree, the MSU faculty average salaries were generally comparable to their regional counterparts. For faculty with a doctorate degree, MSU faculty average salaries were generally less at the Associate Professor and Professor ranks.Generally, the Board of Regents’ Policies and Procedures on Appointment, Promotion, and Tenure is followed to some degree. However, there are differences in how the guidance is applied and the processes employed for salary setting.Key Recommendations Following are the key recommendations in support of the above findings:Make this report available to all MSU tenured and tenure track faculty, as well as those with a role in the compensation process.Conduct further analysis to determine the relationship between salary, school, college, and discipline.Conduct facilitated discussions with faculty to identify issues that are contributing to perceptions of gender-based salary inequity and develop strategies to address these issues.Update, align, and make broadly internally (e.g., posted to MSU’s internal website) available the Board of Regents’ Policies and Procedures on Appointment, Promotion and Tenure and the Faculty Handbook to create a comprehensive resource on MSU talent management.Design a compensation process to compare faculty salaries to appropriate external sources on a fixed basis to ensure competitiveness and equity in salary determinations.Provide annual briefings on compensation policies and procedures, to include relevant and appropriate updates on any compensation initiatives to foster open communication.Charter a faculty compensation working group to address ongoing salary issues and facilitate transparency and open communication.Questions Please direct any questions regarding this study to:Patrina M. Clark, President, Pivotal Practices Consulting LLC6301 Ivy Lane, Suite 108Greenbelt, Maryland 20770(301) 220-3179 officepatrina@IntroductionMorgan State University (MSU), a State-supported university located in a residential area of Baltimore, Maryland, identified a need for assistance from a qualified firm to analyze the salaries among its tenured and tenure track faculty. MSU contracted under Project Number 14/PRO-2008-S (E-Maryland Solicitation MDR 1331011071) with Pivotal Practices Consulting LLC (PPC), a Maryland-based management consulting firm, to:Conduct a study to determine whether there are gender-based salary inequities.Make recommendations to resolve any such inequities.BackgroundMSU’s faculty members deliver a comprehensive set of undergraduate academic programs and a selective set of master’s and doctoral programs to approximately 8,000 students, approximately 70% of whom are Maryland residents. MSU has a total of 43 academic departments listed in their departmental directory. These 43 departments offer:45 programs leading to the bachelor’s degree,30 programs leading to the master’s degree, and15 programs leading to the doctorate, including the online Community College Leadership Program.As a testament to the desirability of MSU’s academic programs, the University has seen a 25% increase in enrollment over the last decade.To remain competitive for the most talented staff available and sustain quality academic programs, MSU recognizes the importance of ensuring pay parity among similarly situated faculty members and acknowledged the following:Salary inequities among similarly situated professionals are disruptive to quality education programs and cause discord within productivity, departments, faculty turnover, and productivity.There is a meaningful distinction between salary differences (based on variances in productivity, contributions to the University, and economic climate) and salary inequities (based on unjustified factors, such as qualifications, assigned responsibilities, and market forces).While no system can perfectly determine to the individual level, the appropriate salary, certain inequities are obvious and extreme as to be readily identifiable and to demand attention and remedy.MSU further acknowledged that identifying internal salary inequities can be difficult because of the large number of variables that can influence pay determinations. Additional potential factors that increase the challenges of identifying these inequities include the absence of:Documentation for setting base salaries, Guidance for awarding incremental salary adjustments, Guidance for awarding discretionary salary increases,Clearly defined pay categories for departments and occupational groups (i.e., position titles).Further, inconsistent application of any established guidance can also be a complicating factor.This study endeavored, to the maximum extent possible, to identify true gender-based inequities based on appropriate statistical analyses.Study Methodology PPC developed a study project plan consistent with widely accepted practices for conducting academic pay studies for higher education faculty identifying critical milestones, collaboration and input points to ensure timely delivery of the study.PPC participated in the following meetings associated with the study:Work initiation meeting on April 8, 2014, with Mrs. Armada Grant, Director of Human Resources, and Mr. Hambisa Belina, MSU Procurement. Meeting with Dr. David Wilson, President, and Mrs. Armada Grant on April 18, 2014.Provided a project overview and status update and addressed questions during the MSU Deans’ Meeting on May 12, 2014.PPC conducted quantitative and qualitative data analysis that included the following elements:Policy and Procedures AnalysisAn internal survey regarding use of internal pay-setting and compensation policies and practices to determine how such practices are applied and if the application of said practices is a contributing factor to gender inequity in salaries for professors, assistant professors and associate professors. A review of MSU’s current written compensation and pay-setting practices to determine how such practices are applied and if the application of documentation is, in any way, a contributing factor to gender inequity in salaries for professors, assistant professors and associate professors. Statistical AnalysisA general statistical analysis of educational compensation to establish the overall context of MSU compensation, policies and pay-setting practices.The development of a comprehensive regression analysis to identify variables that may be impacting gender-based salary decisions. Recommendations were developed based on the results of the application of the study’s methodology.Policy AnalysisInternal Policies and Practices SurveyAn internal survey was conducted regarding the application of compensation policies by college deans and other salary determination participants and stakeholders. The survey was designed to identify any particular gaps or trends that may influence gender based salary decisions. The survey was viewed as an opportunity to identify compensation best practices and processes within MSU, and to assess the faculty’s awareness of MSU’s requirements for salary determinations and merit increases.The survey was sent to 33 participants, and all 33 participants at least partially completed the survey. Of the 33 respondents, 6 respondents provided only their names and titles and the other 27 answered some or all of the questions, resulting in a 100% response rate and a 82% substantive response rate.In summary, PPC did not observe any policy or procedure that resulted in gender-based pay setting or merit pay decision. However, it is noteworthy that several survey respondents identified a lack of replicable written process documentation, which makes it impossible to establish a clear, consistent process that can actually be assessed. In terms of best practices, more than one respondent identified that the process of securing the Dean’s recommendation for submission to the Provost worked well. Finally, there were several recommendations for improving the process, to include establishing a fair, equitable and transparent process that allows for some flexibility.Table 1 on the following page provides an overview of the survey and summarizes the responses and findings. Appendix B includes the narrative comments for process improvement recommendations without individual attribution.Table 1: Internal Policies and Practices Survey SummarySurvey QuestionPurposeResponse/ Finding1. Please provide your name:Self Evident2. What is the title of your current position?Self Evident3. Please describe how you use the "Board of Regents' Policies and Procedures on Appointment, Promotion, and Tenure" and the "Faculty Handbook" in pay setting determinations.To determine consistency in application and knowledge of processes to be used in pay setting.At a variety of levels, respondents indicated awareness that these two documents drive MSU pay-setting processes was evident.4. Please describe the process for setting salary for a new faculty member, to include any peer and/or upper level reviews.To determine consistency in application and knowledge of processes to be used in pay setting for new faculty.All respondents indicated a general understanding of the requirements, with process variations among schools.5. Is this process documented in writing?To assess the existence of written process guidance for the pay setting decision-making process.The process for setting salary of new faculty members is not consistently documented. Nine (9) respondents indicated the process was documented in writing and fourteen (14) indicated it was not.6. Is the written process available to all faculty within the Department and/or School?To assess the degree of transparency regarding the pay setting process.Where written process documentation exists, it is generally made available. Of the nine (9) respondents indicating the process is documented in writing in Question 5, seven (7) responded that it is available to faculty within the Department. Interestingly, two (2) of the faculty members who responded that the process is not documented in writing responded that it is available to faculty.7. Please describe any required review process before a merit salary increase is approved/finalized. If no review process exists, please enter "None."To determine consistency in application of processes to be used in requests for merit salary increases.Almost all respondents indicated the annual report is a key element of the process. Some schools described a fairly robust process for merit increases, with almost all requiring multiple levels of review.8. What aspects of the pay setting process work particularly well?Provide an opportunity to MSU faculty and staff to identify internal best practicesResponses to this item were somewhat sparse, with some indicating a clear need for more robust process rigor beginning with a documented policy.9. What recommendations do you have for improving the pay setting process?Provide an opportunity to MSU faculty and staff to provide process improvement recommendations based on process experienceResponses were more robust than those for Question 8. Responses indicated consensus in the need for a fair, equitable and transparent process that allows for appropriate, supportable flexibilities.Internal Policies and Practices Documentation ReviewAs part of the analysis regarding MSU’s pay practices, a review of the University’s current written compensation and pay-setting practices was conducted to determine how such practices are applied and if the application of the documentation is a contributing factor to gender inequity in salaries for professors, assistant professors and associate professors. The following information was requested:Compensation policies, memorandum, negotiated agreements, processes and any other documentation used for salary determinations. This includes any information MSU may have on classification or positional salary decisions, as well as information on longevity increases, if this salary approach is used by the University.Pay grade matrix for position categories that includes the complete salary ranges (i.e., minimum salary and maximum salary) for each position category.Job analysis for each position category that describes the required knowledge, skills and abilities.Pay setting decision matrix for newly hired faculty members.A title comparability matrix showing how position title variations align with the position categories of #3 above.A description of the bases for discretionary salary increases (e.g., attainment of a terminal degree, publication in nationally recognized trade journal, etc.).Any compensation or pay-related studies conducted by or on behalf of MSU.Copies (may be redacted of personal information) of any compensation complaints or appeals filed under negotiated agreements.While not all requested documentation was received, sufficient information regarding the University’s compensation and pay practices was provided. Analysis focused on the Board of Regents' Policies and Procedures on Appointment, Promotion, and Tenure and the MSU Faculty Handbook. The analysis indicated the following:Board of Regents’ Policies and Procedures on Appointment, Promotion and Tenure: A review of this document did not find any obvious policies or procedures that would result in gender-based pay decisions. All documentation presented is neutral and merit-based. MSU Faculty Handbook: A review of this document did not find any obvious policies or procedures that would result in gender-based pay decisions. All documentation presented is neutral and merit-based. In summary, the documentation review did not identify any policy applications or procedures that would result in a gender-based salary or merit pay decision. There may be some process discrepancies between these two documents, which should be addressed to ensure consistency.Summary of ResultsThe policy analysis and internal policy survey did not reveal any specific trends, gaps or issues that would indicate the MSU compensation and pay-setting policies and practices contribute, either positively or negatively, to gender-based salary decisions. Statistical AnalysisComparison of MSU Faculty Gender Counts and Salaries to National AveragesAccording to the U.S. Department of Education, in 2010-2011 there were 2,870 Title IV post-secondary four-year colleges. The Title IV status encompasses those institutions that have written agreements with the Secretary of Education to participate in Title IV federal student financial assistance programs. Within the state of Maryland, this includes a total of 62 institutions, 30 of which are public. Comparison of Salary AveragesTo establish a baseline for the analysis, current average MSU salaries were compared to the national averages for public institutions, differentiating for educational level. The following two tables provide comparisons of the national and MSU average faculty salaries and include differences in dollars and percentages. Specifically:Table 2 provides a comparison of faculty with doctorate degrees.Table 3 provides a comparison of faculty with master’s degrees.Table 2: Education Level Doctorate, National Public Institutions ComparisonFor faculty with doctorate degrees, the average salaries for MSU faculty as compared to their national counterparts are:Less than the national averages at the Associate Professor and Professor ranks; and More than the national average at the Assistant Professor rank. Gender variance is most pronounced at the Assistant Professor level, where MSU females’ average salary is 14.96% higher than the national average, while MSU males’ average salary is less than 1% higher.Table 3: Education Level Master’s, National Public Institutions ComparisonFor faculty with master’s degrees, the average salaries for MSU faculty as compared to their national counterparts are generally equal. Again, MSU female Assistant Professors earned more (5.2%) than their national counterparts. Data is not provided for the Professor rank in Table 2 because MSU faculty at the professor rank was reported to have doctorate degrees.From a regional perspective, Tables 4 and 5 provide a comparison of faculty salaries as compared to public institutions within the South Atlantic, which comprises the states of Delaware, District of Columbia, Florida, Georgia, Maryland, North Carolina, Puerto Rico, South Carolina, Virgin Islands, Virginia, and West Virginia. These comparisons are gender neutral and show the differences in dollars and by percentage.Table 4: Education Level Doctorate, Regional Public Institution Salary ComparisonFor faculty with doctorate degrees, average salaries for MSU faculty are slightly above regional counterparts at the Assistant Professor rank. At the Associate Professor and Professor ranks, however, salaries are below regional counterparts (6.3% less at the Associate Professor rank and 29% less at the Professor rank). Table 5: Education Level Master’s, Regional Public Institution Salary ComparisonFor MSU faculty with master’s degrees, average salaries were comparable to those of their regional counterparts. Assistant Professors earned 1.1% more and Associate Professors earned 1.5% more than their regional counterparts. Again, there is no MSU faculty at the Professor rank holding master’s position of the Faculty PoolIn reviewing average salaries, consideration must also be given to the percentages of males and females in the faculty pool. The National Center for Education Statistics (NCES) is the primary federal entity for collecting and analyzing data related to education in the United States and other nations. NCES fulfills a Congressional mandate to collect, collate, analyze, and report complete statistics on the condition of American education; conduct and publish reports; and review and report on education activities internationally.Figure 1 provides a comparison of faculty by gender, as a percentage, at the national level in 2011 as reported in NCES’s fall digest. Figure SEQ Figure \* ARABIC 1: Post-Secondary Faculty by Gender, National TotalsComparatively, Figure 2 reflects faculty by gender, as a percentage, at MSU based on the faculty data provided on April 17 and 18, 2014 (see next page).Figure SEQ Figure \* ARABIC 2: Faculty by Gender (2014), MSUIn comparing the national and MSU data, the percentages for MSU were higher than the national percentages for males in Professor (75% MSU versus 70.88% nationally) and Associate Professor (62.73% MSU versus 57.82% nationally) faculty positions. The only faculty position at MSU for which females constitute a higher percentage than the national average is the Assistant Professor (51.9% MSU versus 49.34% nationally). The overall gender representation for all faculty positions nationally as compared to MSU was relatively consistent with the total percentage of females being only slightly higher at MSU (40.72% MSU versus 39.98% nationally). Data Analysis and MethodologyData analysis was conducted to determine, among other things, the impact gender had on salary. Additionally, the analysis took into account other data considered relevant to salary determination, including tenure status, years of service, and the school/discipline in which the faculty was associated. The analysis conducted was based on demographic and transactional (e.g., salary increases for cost-of-living adjustments, promotions, and merit) data. The transactional data covered salary adjustments for the period August 2012 through April 2014. Because of differences in certain fields in the data provided, data was merged using the VLOOKUP function in Microsoft? Excel. Specifically, this was done using the faculty ID numbers to locate information in other data sources. Additionally, one anomaly, namely a faculty member with the rank title of “Professor” whose salary ($344,597) was significantly above the average salary for professors, was removed so as not to influence the analysis results.As an initial analysis, data was reviewed to look at the overall number of employees by gender and rank (Assistant and Associate Professor and Professor), including average salaries, average years of service, and tenure status. Relationships among the variables to look for indications of gender bias were also reviewed. Following this analysis, regression analysis was performed based on overall faculty within the three ranks in addition to separate analyses for each of the faculty rankings. Subsequent to the regression, an analysis was conducted that examined transactional data, and in particular salary increase rates, promotions, and new hires. Assumptions and Data LimitationsIn conducting this analysis, the following assumptions were made:Because all faculty within the three ranks hold either masters or doctorate degrees, and based on the principle of parsimony, education levels were not considered in the regression analysis.The data provided did not contain fields related to promotion rationale and was limited to two years (2012 and 2013). There were, however, a limited number of promotions (6 in 2012 and 17 in 2013), which limits the ability to conduct statistical analysis.While some new hire data was provided, additional factors, such as previous years of experience, was not available. Therefore, the impact of these factors was not addressed in this study.Demographic Data AnalysisBefore conducting regression analysis, it is often beneficial to conduct a high-level analysis to identify any discrepancies and better focus the regression analysis. Common elements of this high-level analysis include ratios and averages.Toward that end, we established average salaries by rank (Assistant Professor, Associate Professor, and Professor) as well as minimum and maximum salaries, standard deviations, and expected minimum and maximum salaries given the calculated standard deviations. Table 6 below provides the outcome of this analysis.Table 6: Average Salaries, Standard Deviation, and Expected Salaries - By RankBased on the calculated expected values, the data was further analyzed to determine the number of faculty, by rank and gender, whose salaries fell outside the expected range. Table 6 presents those results.Table 7: Salaries, by Rank and Gender, Outside Expected RangeAs the data in Table 7 shows, 22.4% of all faculty at the Assistant Professor rank have salaries that fall outside (either above or below) the expected range. This percentage increases to 35.8% for Associate Professors, and 58.1% for Professors. In terms of gender, 29.3% of all female Assistant Professors have salaries outside the range (as compares to 14.3% of males), while 32.5% of female Associate Professors (compared to 37.7% of their male counterparts), and 25.0% of female Professors (compared to 30.4% of their male counterparts) have salaries that fall outside the expected ranges. Univariate, Bivariate, and Multivariate AnalysesOther than salary, data on other aspects of the faculty including years of service, tenure, and affiliated schools (disciplines) were also provided. This data was reviewed to identify any potential relationships among the various demographic factors and, more specifically, as relates to gender.Tenure StatusTable 8 reflects the relationship of tenured to non-tenured faculty by both rank and gender. Specifically, there are a total of 131 tenured faculty within the Assistant, Associate, and Professor ranks. Of these 131 faculty, 46 (35.1%) are female. Additionally regarding gender, the majority of both female (73.9%) and male (65.9%) tenured faculty are at the Associate Professor rank. Table 8: Number and Percentage of Tenured and Non-Tenured Faculty, by GenderAverage Years of ServiceTable 9 provides the average years of service by both rank and gender. At the Assistant and Associate Professor ranks, females have fewer average years of service than their male counterparts. At the Professor level, however, the average years of service for females is slightly higher than for male Professors.Table 9: Average Years of Service, by Rank and GenderTenure and Average Years of ServiceNon-tenured females at all faculty ranks averaged lower years of service than their male counterparts. From a tenure perspective, female Professors averaged slightly higher average years of service than their male counterparts. Table 10 provides a comparison by rank and gender, of average years of service at both the tenured and non-tenured levels.Table 10: Tenure and Average Years of Service, by Rank and GenderFaculty by College (Discipline)In reviewing the school/discipline in which faculty is associated with average salaries (Tables 11 and 12), female faculty are most concentrated in Business and Liberal Arts. From a salary perspective, average salaries for females were also higher than their male counterparts in those disciplines at all levels with the exception of Professor (the average female salary was less in the Business discipline).In the Engineering discipline, female faculty averaged lower salaries than their male counterparts at all rank levels. Alternatively, in both the Health and Liberal disciplines, female faculty averaged higher salaries than their male counterparts at all rank levels. Table 11: MSU Faculty by College, Rank, and GenderTable 12: Average Annual Salary by College (Discipline), Rank, and GenderFinally, in comparing student enrollment (2006 figures, depicted in Table 13 below) to faculty data, while there is a greater number of faculty, both male and female, in those schools/disciplines in which there is higher enrollment, the impact of enrollment on average salaries is not consistent. For example, in the Business discipline, female Assistant Professors had an average salary of $114,804.91 as compared to an average salary of $60,055.92 for their female counterparts in the Liberal discipline. (Note: 2006 figures are the most recent available for this analysis – see footnote for sourcing.) Table 13: MSU Enrollment by School/College, 2006 Business840 695 1,535 25.2% Education418 186 604 9.9% Engineering213 522 735 12.1% Health70 20 90 1.5% Liberal1,184 673 1,857 30.5% SCMNS438 337 775 12.7% Architecture114 144 258 4.2% Social Work203 41 244 4.0%Total:3,480 2,618 6,098 100.0%FemaleMaleTotal% TotalSchool2006 Enrollment Business840 695 1,535 25.2% Education418 186 604 9.9% Engineering213 522 735 12.1% Health70 20 90 1.5% Liberal1,184 673 1,857 30.5% SCMNS438 337 775 12.7% Architecture114 144 258 4.2% Social Work203 41 244 4.0%Total:3,480 2,618 6,098 100.0%FemaleMaleTotal% TotalSchool2006 EnrollmentRegression Analysis - OverviewRegression analysis is used to study the relationship among variables; specifically, the impact of independent, or explanatory, variables on a single dependent, or response, variable. In the case of this study, the dependent variable is salary. Independent variables consisted of those factors considered most likely to have impact on faculty salaries, namely years of service, tenure status, and the school in which the faculty member is affiliated. Regression explains the fit of the line through the independent variables; the closer the fit, the greater the degree of explanation (i.e., how much of the dependent variable can be explained by the independent variable(s)).While the output of regression analysis includes several statistics, those most frequently analyzed (and reviewed for this study) are as follows:R-Square: the “coefficient of determination,” which is always between 0 and 1, and explains the percentage of variation of the response or independent variable(s) explained by the regression line. Adjusted R-Square: the R-Square value, adjusted for testing multiple variables.Standard Error (of Estimate): indicates the typical error likely to be made using the fitted value based on the regression line.Coefficient: the numerical value of the constant (in this study, the constant is the expected salary) and independent variables (either positive or negative numbers).T-Value: ratio of the estimated coefficient to its standard error, indicating the number of standard errors the regression coefficient is from 0. T-values are used for hypothesis testing to determine if the regression coefficient belongs in the regression equation. Where the t-value is <1, the Adjusted R-Square value will increase.P-Value: the probability a value will be beyond the t-value. Parsimony is a term that means explaining the most with the least. For example, if two independent variables yield an R-Square value of 0.980 while the use of 10 independent variables yield an R-Square value of 0.981, then the parsimony principle indicates that only the two variables should be used. Regression Analysis – ResultsThe analysis for the MSU salary study was conducted in a tiered manner, first looking at faculty as a whole, followed by the three groups of Assistant Professors, Associate Professors, and Professors. This tiering of the data was done based on the varied results from the univariate, bivariate, and multivariate analyses conducted above. Additionally, the regression was done initially as simple regression based on the relationship between gender and salary. Subsequently, and based on the initial results, other factors were added to examine what relationship, if any, existed between those additional factors and MSU faculty salaries.In terms of overall findings, it was evident from the demographic data analysis that there were wide disparities in salary data among the various faculty ranks and genders. Likewise, the resulting regression analysis showed very high P-values in many instances, indicating that most of the variables should be excluded. However, continuing the regression by adding other independent variables ultimately increased adjusted R-squared values to more than 50% in most instances. Although 50% is well below the threshold of what is normally considered explanative (typically, adjusted R-squared values should be at the 85% level or higher), the results indicated that the factors analyzed explained approximately half of the differences in salaries across the organization. While the resulting formulas are of very little value in predicting salaries, they are indicative that faculty salary structure is widely variable, and gender alone provides no explanation as to salary variances.As an initial regression, gender was analyzed (Table 13). In all but one instance (Associate Professor rank), however, the adjusted R-square value was negative and, in general, the adjusted R-square values were less than 2% for all ranks. This is indicative that gender was NOT predictive of salary for any of the ranks. Based on these results, the resulting formula for a female’s salary is as follows:Assistant Professor: $71,299 + $7,994 = $79,293 (error margin of $5,094)Associate Professor: $84,907 - $2,957 = $ 81,950 (error margin of $4,068)Professor: $101,587 + $6,167 = $107,754 (error margin of $9,025)In these formulas, the base salary, that ranges from $71,299 for Assistant Professors to $101,587 for Professors, is adjusted in a positive direction at both the Assistant Professor and Professor ranks but is adjusted downwards at the Associate Professor rank. The error margins indicate the variation for each rank, e.g., female Assistant Professor salaries range from $74,199 to 84,387 based on the resulting formula.Table 14: Regression Results - GenderSubsequently, the regression independent variables were expanded to include both gender and tenure. Results are provided in Table 15.Table 15: Regression Results - Gender and TenureNotable from this analysis is that the rank of Professor results in an adjusted R-square value of 15.6%, indicating that the factors of gender and tenure explain 15.6% of a Professor’s salary. While tenure is still of very little value in predicting salary, it does have some significance at the Professor level; however, the relationship to salary is negative. Without considering the applicable error margins, for a female with tenure, the resulting salary prediction formulas are as follows:Assistant Professor: $73,145 + $6,936 – $8,073 = $72,008Associate Professor: $79,999 - $3,207 +$6,068 =$82,860Professor: $136,573 + $9,666 - $38,485 = $107,754Years of Service was added as a variable for the third regression analysis. Consistent with the findings for gender and tenure, this variable had minimal impact on salary. For Assistant Professor and Professor ranks, it actually decreased the R-squared values. Table 16 provides the overall results.Table 16: Regression Results - Gender, Tenure, and Years of ServiceThe faculty member’s school/college was added as a variable in the final regression, the results of which are shown in Table 17. This variable increased the adjusted R-square value for both Assistant and Associate Professor to 66.7% and 56.2%, respectively; meaning that 66.7% of Assistant Professor salaries and 56.2% of Associate Professor salaries can be explained by the variables of gender, tenure, years of service, and school. At the Professor level, however, the adjusted R-square value decreased to 8.3%. Overall, coefficients for the various schools/disciplines resulted in both positive and negative values, ranging from a premium of $29,701 for faculty in the School of Business to a reduction of -$13,622 for faculty in the School of Social Work. Using the base salary (Constant) of $79,047, faculty in the School of Business, notwithstanding the factors of gender, tenure, and years of service, would earn $108,748 ($79,047 + $29,701), while faculty in the School of Social Work would earn $65,425 ($79,047 - $13,622). Table 17: Regression Results - Gender, Tenure, Years of Service, and SchoolTransactional Data AnalysisSubsequent to the analysis of demographic data, an analysis of the transactional data was undertaken to look for other trends or relationships evident between salaries and gender (female) and to determine if additional regression analyses was merited. Although there were significantly more data provided for transactional events, it encompassed only a three-year span (from 2012 to 2014).Table 18 depicts new hires during 2012 and 2013 by rank, gender, and average salary. Females constituted more than 58% of all new hires, with 83% hired at the Assistant Professor level and 17% at the Associate Professor level. Males constituted approximately 42% of new hires, with 71% hired at the Assistant Professor level and 24% at the Associate Professor level. An equal number of males and females were hired at the Associate Professor level, and the average salary for males was less than 1% higher than their female counterparts at this level. Table 18: Number and Average Salaries of New Hires, by Rank and GenderRankNew HiresFemaleMale#Avg. Salary#Avg. SalaryAssistant Professor20 $ 83,830.90 12 $ 69,503.75 Associate Professor4 $ 78,715.75 4 $ 78,874.25 Professor0 $ - 1 $ 115,566.00 Total/Average:24 $ 81,273.33 17 $ 87,981.33 In terms of starting salaries, female Assistant Professor new hires earned on average 20.61% more than their male counterparts. At the Associate Professor level, however, females earned slightly less (0.20%) than their male counterparts.Table 19 below provides a breakout of these promotions by gender. As a second analysis, and from a promotions perspective, while there were relatively few promotions in 2012 and 2013 (6 and 17 respectively), females accounted for more promotions at the Assistant Professor and Associate Professor levels in 2012 and 2013. Of the total seven promotions from Associate Professor to Professor in 2013, two (22.2%) were for female faculty and five (62.5%) were for male faculty. The only promotion from Associate Professor to Professor in 2012 was for a female faculty member.In reviewing average salary increases due to promotion as a percentage of the average salaries for each level, the average increase at the Associate Professor level for female faculty was 1.4% and for male faculty was 2.2%. At the Professor level, the average salary increase associated with a promotion was 2.6% for female faculty and 2.5% for male faculty. This does not vary from a noteworthy perspective from either national or regional comparison.Table 19: Number of Promotions by Gender and Effective Date (Year)As a third and final analysis, other salary adjustments were reviewed. These adjustments included variables other than promotion, i.e., cost-of-living adjustments (COLAs), merit raises, and pay increases. In general, fewer female faculty were granted increases than male faculty. Specifically, female faculty accounted for 196, or 42.2%, of all COLA increases, while males accounted for 269, or 57.9%. In terms of merit increases, females accounted for 75, or 41%, while males accounted for 108, or 59%. In reviewing average increases, female faculty received slightly higher COLA adjustments at the Assistant Professor level by a full percentage point (3.4% female versus 2.4% male), but COLA increases at the Associate Professor level were more than a full percentage point less (2.4% female versus 3.6% male). At the Professor level, female faculty received greater salary increases for COLAs (5.1% female versus 2.4% male), as well as merit increases (3.4% female versus 3.0% male). Table 20 provides further detail on these results.Table 20: Adjustments by Type, Number, and Average % IncreaseSummary of ResultsThis study was undertaken specifically to examine gender disparities in the area of faculty salaries, specifically at the levels of Assistant Professor, Associate Professor, and Professor. Based on a review and analysis of the data, the following provides a summary of results:While there are significant disparities in salaries among MSU faculty, there is no evidence that salary disparities are based on gender bias. In performing regression analysis based only on salary and gender, the adjusted R-square was both negative and less than 0.5%, which indicates no relationship between gender and salary exists.In comparing MSU faculty with averages at the national level, while female faculty at the Assistant Professor level average salaries significantly higher than the national average, this gap does not continue at the Associate Professor and Professor ranks. However, the salary disparity between male and female faculty at the higher levels, and in particular the Professor level, is less than the gap at the national average level.At a regional level, while data specifically on gender is not available, MSU faculty salaries were generally comparable to their regional counterparts for those with master’s degrees. For faculty holding doctorate degrees, however, average salaries were less at both the Associate Professor and Professor ranks.While the regression analysis did not show any meaningful relationships among gender, tenure, years of service, and salary, it did reveal there may be some significance in the school/college/discipline a faculty member is associated with and salary. The relationship, however, is negative for certain schools, i.e., liberal arts. In comparing student enrollment with faculty ranks, the area of liberal arts represented the highest number of students enrolled yet the average faculty salaries, both male and female, were less than average salaries, and in most cases significantly so. This suggests that salary differences may be more influenced by market forces, i.e., the supply of faculty in various disciplines, than by other demographic factors.Reviewing other data, including new hire and promotions data, revealed some interesting relationships among gender and salaries. While female faculty receives significantly higher salaries at the Assistant Professor level, this gap is eliminated at the Associate Professor and Professor levels. Female faculty accounted for more promotions in 2013 (9 female promotions versus 8 male promotions), although average salary increases due to promotions were slightly less for females at the Associate Professor level (1.4% females versus 2.2% males) and slightly higher at the Professor level (2.6% females versus 2.5% males).Since the regression analysis performed reflected that the school/college/discipline in which faculty is associated has some significance on salary, areas of further analysis may include: Comparison of average faculty salaries by degree (master’s, doctorate) and discipline area (e.g., business, liberal arts) to estimates of actual and future employability of those job markets, including predicted expansions, contractions, and changes in overall skill sets and knowledge parison of academic disciplines (e.g., science, engineering) in which females have historically been under-represented with current parison of academic disciplines (e.g., health, liberal arts), in which females have historically been the dominant gender with current faculty.There is some evidence provided in literature reviews, both anecdotal and peer-reviewed, that suggest gender biases exist in certain occupations, which relate to educational disciplines that have traditionally been male dominated. Additionally, certain female-dominated occupations have been historically devalued, resulting in lower overall salaries. Additional analysis in this area may reveal useful information on economic and market forces and their influence on gender salaries.Recommendations As indicated in the summary of results, the study did not identify any MSU policies or practices that would influence gender-based salary decisions.Engage faculty in a facilitated discussion to surface specific points of contention regarding gender-based pay perceptions and identify solutions to address these perceptions. Small focus groups will allow MSU to constructively engage faculty in identify issues for further exploration and appropriate solutions.Conduct briefings on this report at appropriate staff briefings to leverage the findings regarding gender-based salary determinations. Design a process to compare faculty salaries to appropriate external sources on a fixed basis to ensure competitiveness and equity in salary determinations.Establish a committee to update and align the MSU Board of Regents’ Policies and Procedures on Appointment, Promotion and Tenure and the MSU Faculty Handbook to create a “one-stop” information source for decision-makers and faculty members regarding pay, performance, and other resource management issues.Provide annual updates on current MSU compensation policies and procedures to all faculty and staff who are involved in the compensation process.Since the performed regression analysis reflected that the school/college/discipline in which faculty is associated has some significance on salary, areas of further analysis may include and are recommended: Comparison of average faculty salaries by degree (masters, doctorate) and discipline area (e.g., business, liberal arts) to estimates of actual and future employability of those job markets, including predicted expansions, contractions, and changes in overall skill sets and knowledge parison of academic disciplines (e.g., science, engineering) in which females have historically been under-represented with current parison of academic disciplines (e.g., health, liberal arts) in which females have historically been the dominant gender with current faculty.Salary Compression While a review of salary compression is beyond the scope of this study, MSU has expressed an interest in understanding whether salary compression exists and its potential impact on MSU faculty salaries.? The following information on salary compression is provided at the request of the University.Salary compression occurs when there is only a small difference in pay between employees regardless of their skills or experience. It is the result of the market rate for a given job outpacing the increases historically given by the organization to high-tenure employees. Therefore, newcomers can only be recruited by offering them as much or more than senior professionals.According to an article in Inside Higher Ed, it is difficult to tell how widespread a problem salary compression is for two primary reasons:Pay for existing faculty tends to increase over time at the rate of inflation.Pay for new faculty is usually negotiated based on the market rate. While salary compression can be a faculty morale issue that should be addressed, market realities seem to create and sustain a certain amount of compression.When addressing salary compression, an institution must thoughtfully identify solutions that can be implemented within the existing culture and any budgetary constraints.Appendix A: ReferencesU.S. Department of Education, National Center for Education Statistics. (2013). Digest of Education Statistics, 2012 (NCES 2014-015), Chapter 2. Found at: Center for Education Statistics (NCES), data from the Integrated Postsecondary Education Data System (IPEDS), 2009 through 2011. Found at: Wooten, B. H. Gender Differences in Occupational Occupations. US Department of Census, Bureau of Labor Statistics (BLS), Monthly Labor Review. April 1997. Found at: NCES, Digest of Education Statistics, Full-time instructional faculty in degree-granting institutions, by race/ethnicity, sex, and academic rank: Fall 2007, Fall 2009, and Fall 2011. Found at: MSU 2006 Enrollment by School/College, found at: Mangan, K. Despite Efforts to Close Gender Gaps, Some Disciplines Remain Lopsided. The Chronicle of Higher Education. October 29, 2012. Found at: The Simple Truth About Gender Pay Gap. American Association of University Women (AAUW), March 10, 2014. Found at: Bellas, ML. Disciplinary Differences in Faculty Salaries: Does Gender Play a Role? The Journal of Higher Education, Vol. 68, No 3.Appendix B: Faculty and Staff Recommendations for Process ImprovementQ9. What recommendations do you have for improving the pay setting process?The process should be reviewed more often to keep pay current and competitive.In a market-driven environment, new faculty salaries are higher than more senior faculty who have served the University for many years. Periodic reviews are needed for salary increases and equity. Reviews for salary equity should consider faculty contributions in various areas.Require chairs to more consistently link their grading of their faculty member's performance with a merit pay recommendation so that there is not a wide variance between the merit pay recommendations for faculty members with the same (or nearly the same) performance ratings.Accept the practice used by the School of Business. Other disciplines may have similar databases, which need to be used justifiably.The dean should be able to set the salary offer and not the Provost who is far removed from the discipline or uses websites like "" as his basis. Most of the salaries in that website are not geared towards academic positions.Establishing a process for salary setting within the University that allows for variation across disciplines/schools.Providing sufficient time for faculty to complete annual reports and Chairs to make evaluations for use in the merit process.I would like to become more familiar with the salary settings across disciplines.Establishing a pay range for each faculty rank based on data that is pertinent to the discipline; accepting the recommendation of the Dean who is more knowledgeable about salaries that those who make the decision.Should be based on fairness and equity.Develop a way to "catch up" salaries of existing personnel following a period of salary freezes. There needs to be something other than merit, which can bring all well performing employee salaries up to competitive levels. In our area, faculty salaries are typically 20% below the national norm. Also the admin salaries need to be increased to expect courteous good quality output. These salaries should also be determined by the direct supervisor. Better coordination of the availability of the information used to determine pay increases (FAR data in a usable form).There does need to be a published salary range that is based on rank, years in service, and broad disciplinary engagement. The disparity in salary ranges in the College has very little logic, especially given the disparity in rank and experience. I think that we need to have a salary range based on rank that is university wide first, and then disciplinary focused second. For example, all full professors should have a range of $85,000 - $110,000 (a guestimate), and then consider disciplines. The salary should take into account publications/creativity, grant writing and fund raising, disciplinary participation and service, major administrative service, and community service and engagement. There should be a published standard for each rank.It would be useful if the response to the recommendations made by the Deans and Chairs were relay[ed] back to them.No comment.I do not know what the pay setting process is. More transparency in the process is surely desirable.Gender should not have an impact on salary.It should involve the faculty and the chairperson and the Dean. Once the recommendation is done, it should not be changed in the Office of the Provost. In the most recent merit increase, an individual who was given a zero increase at the departmental level for lack of productivity in research and teaching effectiveness ended up with a 5% increase when the final merit increases were announced. This shouldn't have happened that way. Thanks.White women at Morgan State University appear targeted for the lowest salaries. Within my school, the lowest paid individuals if compared relative to the criteria of 1) length of service; 2) prior experience in field: 3) quality of educational preparation; 4) distinctions such as service as Senior Fulbright Scholars (for example) are the lowest paid. Then next lowest paid cohort is the white male registered architect, because the university appears to only give "lip service", or false assertions of parity between professional degrees, professional registration, and academic terminal degrees such as the Ph.D. The university administration does not have the knowledge that architecture is not the same as engineering in terms of final degrees, and other factors. My recommendation is that for professional schools and fields, the university should look at national standards of pay setting for professionals within universities. Next, existing and dramatic discrepancies between individual salaries should be resolved. White women are the most dramatically underpaid, based on my knowledge of salaries at my school.Give the form earlier in the process.None.Base salary scales on rank and similar other to state institutions.Criteria should be applied and shared among administration and possibly include the faculty as needed. Transparency at all stages would be helpful.Appendix C: Regression Analysis Summary Output TablesGENDERALLSUMMARY OUTPUTRegression StatisticsMultiple R0.030978455R Square0.000959665Adjusted R Square-0.00370875Standard Error23067.1366Observations216ANOVAdfSSMSFSignificance FRegression1109379928.1109379928.10.2055655140.650724937Residual2141.13868E+11532092790.8Total2151.13977E+11CoefficientsStandard Errort StatP-valueLower 95%Upper 95%Intercept84491.173232046.87738141.278082417.1457E-10480456.5501888525.79628GENDER-1445.768733188.773698-0.453393330.650724937-7731.196474839.659Expected Salary-Female:83,045.40$ Expected Salary-Male:84,491.17$ Variance (%)-1.7111%ALLSUMMARY OUTPUTRegression StatisticsMultiple R0.030978455R Square0.000959665Adjusted R Square-0.00370875Standard Error23067.1366Observations216ANOVAdfSSMSFSignificance FRegression1109379928.1109379928.10.2055655140.650724937Residual2141.13868E+11532092790.8Total2151.13977E+11CoefficientsStandard Errort StatP-valueLower 95%Upper 95%Intercept84491.173232046.87738141.278082417.1457E-10480456.5501888525.79628GENDER-1445.768733188.773698-0.453393330.650724937-7731.196474839.659Expected Salary-Female:83,045.40$ Expected Salary-Male:84,491.17$ Variance (%)-1.7111% LINK Excel.Sheet.8 "C:\\Users\\Katie\\Documents\\Custard Consulting\\Regression_5-26-2014.xlsx" "GENDER_All RA!R1C1:R24C11" \a \f 4 \h GENDER AND TENUREGENDER, TENURE, YEARS OF SERVICEGENDER, TENURE, YEARS OF SERVICE, SCHOOL ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download