Merit-based Faculty Compensation

prepared by

Lawford J. Anderson, Joseph Aoun, Michael Appleman, Tu-nan Chang,
Joseph Hellige, and Sarah Pratt

University of Southern California

Executive Summary

I. Roles of Faculty in Changing Times

II. Faculty Activity Profile

1. Definition of activities

2. Distribution of effort

III. Merit Evaluation

1. Evaluation criteria

2. Evaluation processes

IV. Salary Recommendation

V. Other Issues


I. Roles of Faculty in Changing Times

Recent economic and social changes have prompted many people - from government officials to representatives of the media, college students and their parents, university administrators, and faculty members themselves - to call for greater accountability on the part of universities for the "products" and "services" they provide. This demand for greater accountability, combined with changes in the structural environment of higher education, has led to further demands for a reexamination of faculty responsibilities.

As a faculty group, we welcome the notion of accountability. At the same time, we note that concepts of responsibility and performance form a natural partnership with concepts of rights and privileges. The most basic right of faculty members is the right to academic freedom guaranteed by tenure through a thorough and fair process of peer review. Without the right to academic freedom, the notion of faculty responsibilities degenerates into something like a pay-per-piece assembly line mentality inimical to the building of great universities, which are inevitably based on creative and free inquiry carried out by a mutually cooperative community of scholars. Every aspect of this white paper rests on this premise. In addition, we concur with the resolution passed on May 8, 1996 by the USC Academic Senate, which states that "the principles embodied in the tenure system are central to the mission and the very nature of the university."

Given a strong commitment to tenure, we are prepared to take on the issue of faculty responsibility. Among the major structural factors responsible for the uncertainty of the funds available for faculty support that necessitate the current discussion are i) the decreases and changes in the direction of funding from government institutions and other grant providers, ii) the limitation of future tuition increases and potential reduction in student financial aid, and iii) the elimination of mandatory retirement. (A detailed listing of the number of tenured faculty in each five-year age group at USC can be found in last year's Academic Senate white paper on Faculty Rejuvenation and Retirement.)

The obvious solution to these problems is to increase the financial resources as much as possible, and to make the best possible use of the resources that are available. We feel that emphasis should be placed on the following:

Taking the Academic Senate white papers of past two years on Faculty Responsibilities as a starting point, we would like to reiterate a few key passages that read as follows:

Given a strong faculty commitment to academic excellence, a merit-based compensation system developed by elected bodies of faculty governance (or faculty groups appointed by the elected bodies) in conjunction with the administration will help us sustain a long term vitality of our faculty under a fast changing external environment. One of the key ingredients in the establishment of such a system is a strong partnership between the faculty and the university administration, i.e., it should be initiated by the faculty and developed jointly by the faculty and the administration, at both the school and the university level. We are well positioned at USC to meet such a challenge with the current faculty governance structure. It is our hope that this white paper will lead to additional faculty discussions at USC in our attempt to establish a sound and effective faculty compensation system which will ensure the financial security of our faculty as well as the long term fiscal stability of the University.

Three key components of a faculty compensation system, i.e., faculty activities profile, merit evaluation, and salary recommendation, will be discussed in Sections II - IV, respectively. Many recommendations discussed in this white paper have, in fact, already been employed by some of the schools at USC.

II. Faculty Activity Profile

During the past two decades at USC, the most commonly applied instrument for defining the relative proportion of faculty effort devoted to three major categories (i.e., instructional activities, scholarly activities, and service) is the Faculty Activity Profile, formerly known as the Spitzer Profile. Because of the differences among various fields of endeavor, the basic parameters for the nature, quantity, and quality of full-time work are best determined at the school level. These parameters should be set by the elected faculty governance body, or a group appointed by the officers of the elected body, and the cognizant dean. Many academic units at USC have already applied such a procedure to determine the faculty activity profile in accordance with the standards established by their academic unit.

II.1 Definition of Activities

The primary responsibility of a faculty member is to contribute to the mission of the University by engaging in i) high quality scholarly activities which eventually will become the source of new knowledge, technology, or art, ii) instructional activities to disseminate knowledge and help students to acquire wisdom and insight, and iii) constructive faculty service which is not only important but necessary to develop and to maintain a favorable environment conducive to scholarly and instructional activities. The lists below represent our attempt to summarize a possible classification of activities in the three major categories. We recognize that various kinds of academic activities are closely intertwined and believe that the categories should have considerable flexibility.

A. Instructional Activites:

  1. Classroom teaching - includes the presentation of lectures, leading seminars and discussion sections, and directing instructional laboratories in officially listed courses. This would include regularly scheduled general education, undergraduate and graduate courses, but not Directed Research.
  2. Curriculum development - includes the design and implementation of new or remodeled programs, curricula, or courses in general education or new and existing minors/majors.
  3. Academic mentoring and direction - includes academic interactions with graduate or undergraduate students, either individually or in small groups, for the purposes of research guidance (e.g., Directed Research) or dissertation direction involving specific project. (In some units, supervision of students in specific project may more appropriately be classified as a research activity, rather than an instructional activity.)
  4. Advisement - includes meetings with students for purposes of registration, general guidance, counseling, and career planning.
  5. Others - includes other educational or instructional activities (e.g., outreach programs) involving outside institutions. It may also include the development of new teaching materials (e.g., textbook or extended lecture notes) or technology (e.g., use of multimedia approach) which contributes beyond the boundaries of the University.

B. Scholarly activities:

  1. Research - includes the use of literature and the laboratory or field investigation leading to the dissemination of knowledge through peer evaluated original publications, reviews, books, or seminars/talks at professional conferences and other academic institutes.
  2. Production - includes the design, creation, or implementation of works of arts, music, and performance.
  3. Funding - includes the preparation of proposals and/or the acquisition of funding which supports the scholarly works of a faculty member.
  4. Participation in Professional Societies and Conference Organization - includes taking part in the activites of a discipline by serving on peer review panels (e.g., membership on Editorial Board and referee for journals or funding agencies), committees, or advisory boards of professional organizations or research institutes. It may also include the organization or the participation in the organization of professional conferences/workshops.
  5. Others - includes other research related activities, e.g., consulting or development of educational technology which contributes to one's discipline. It may also include directing or mentoring students or postdoctoral fellows, either individually or in small groups involving specific project.

C. Service:

  1. Departmental activities - includes services in an appointed or elected position which contribute to the operation of an academic department (e.g., Department Chair, Associate Chair, or membership in departmental committees).
  2. School and University activities - includes services in appointed administrative positions or membership in faculty committees or task forces at the school or university level.
  3. Faculty Governance - includes service as an appointed or elected member of the Academic Senate or School Faculty Council/Assembly or membership in Academic Senate or School Council/Assembly committees or task forces.
  4. Others - includes outreach activities such as those associated with the USC strategic initiatives and other university endeavors.

II.2 Distribution of effort

A Faculty Activity Profile should be agreed upon mutually between a faculty member and his/her department Chair/Dean in accordance with his/her faculty responsibilities. Since a strong tenure system is established based on a comprehensive peer review of the performance and achievement of the faculty in all three categories of activities listed above, it is expected that he/she will continue to contribute significantly in all three areas. Therefore, in addition to the recommendation of the establishment of a "typical distribution" in each academic unit for most the faculty members, last year's white paper has also suggested that a "minimum level of effort" be required in each category of faculty activities.

Based on the recommendations proposed by last year's Academic Senate white papers on Faculty Responsibilities and Faculty Rejuvenation, we suggest the following:

  1. A typical distribution of effort for a full-time faculty in most schools will be

    35-55% (Instruction) + 35-55% (Scholarship) + 5-15% (Service) (must) = 100%.

    This means that 35 - 55% of the individual's effort would be devoted to instructional activities, 35 - 55% to scholarly activities, and 5 - 15% to service. The total must equal 100% if the individual has a full-time appointment. It is expected that majority of the faculty (e.g., in the College of Letters, Arts, and Sciences) will have a typical profile.

  2. Use of an individualized profile - A faculty member selecting an individualized profile may do so in consultation with, and with the approval of, the chair and cognizant dean of the relevant units, as long as the profile does not fall below the minimum percentage for an individualized profile in any category as defined by the school (see item "3." below) and as long as the total yields 100% if the individual has a full-time appointment.
  3. Minimum percentages for an individualized profile - In the absence of special circumstances, a faculty member should devote at least a minimum percentage of effort to each of the three areas of activities. For example, the minimum permissible percentages might be expressed schematically as:

    Instruction > 20% + Scholarship > 20% + Service > 5% (must) = 100%.

    A school is free to establish its own minimum percentages as long as it is developed by a faculty committee made up of the cognizant dean and the elected faculty council (or a subgroup appointed by the elected faculty council) and in accordance with the academic missions of the school.

  4. Less than full time effort - It was suggested by two of the Academic Senate white papers last year that the voluntary use of a less-than-full-time individualized profile be considered for faculty members working more than half-time, but less than full time. This kind of profile could be combined with an early retirement program involving incentives and an irrevocable commitment to retirement for USC faculty members with a fixed minimum number of years of service. It would allow the total effort be reduced to less than 100% with a proportional salary reduction.

III. Merit Evaluation

The employee job evaluation practiced in most of the industry, business, and government offices is typically carried out by an immediate supervisor or manager with limited input from the employee. In contrast, for faculty in higher education institutes, a peer evaluation based on the sound judgement of an elected faculty panel (e.g., Merit Review Committee) is the only realistic and fair form of merit (i.e., performance/productivity) evaluation. A fair and objective peer evaluation generally works best if it is kept in strict confidence and if a personal working relation is not involved. For a group of faculty working within the same academic unit over a long period of time, it is unrealistic to expect that a peer evaluation system, however elaborate, will be free from "grade inflation" even with a periodic calibration. Given the university's commitment to building on excellence, it may be that a relatively high proportion of faculty is placed in the "outstanding" category. We should not consider this inappropriate as long as the quality and the impact of their works can be assured, first, by a thorough and fair tenure and promotion process and, second, by a rigorous post tenure evaluation based on his/her cumulative accomplishment and continuing short term productivity. The merit evaluation process should also be sufficiently effective to differentiate the level of faculty merit within the same academic unit to ensure a fair and equitable salary distribution.

The individual school is clearly in the best position to determine the general parameters which dictate the evaluation criteria for each of the activities listed in Section II.1 above. These parameters should be set by the elected faculty governance body, or a group appointed by the officers of the elected body, and the cognizant dean. More detailed evaluation criteria, which could vary from discipline to discipline, may be determined by the appropriate individual academic unit in accordance with the general parameters established by the school. Along with research and teaching in regular disciplines, interdisciplinary activities should also be recognized, and emphasized when appropriate, as a key component of a merit-based faculty compensation system.

The application of the evaluation criteria may depend upon an elaborate mechanism based on the sum of numerical scores of a long list of detailed criteria, or a system based on personal judgement after a careful reading of the relevant materials. The pros and cons of both approches abound. In practice, as long as the Merit Review Committee consists of elected faculty members, we believe it is likely to adopt a fair and appropriate mechanism. We are more concerned with the lack of i) willingness of the faculty to serve on the Merit Review Committee and ii) incentive for the Merit Review Committee to put in a real effort, when the size of the salary pool is much too small to distinct any real difference in salary increases.

Following the evaluation, the Merit Review Committee should place each faculty member into a specific "rank," which summarizes i) the short term productivity over the immediate preceeding year or a number of years since his/her most recent evaluation and ii) his/her cumulatvie performance and professional achievement. The number of ranks should be sufficiently large to provide a meaningful discrimination of different levels of faculty merit. We should also keep in mind that in any given system, very few, if any, faculty are likely to be assigned into the lowest rank. A narrative system, based on a detailed summary of the works and their impact, has also been employed in a limited number of disciplines (e.g., in Humanities) and is viewed as a fair and effective evaluation method by the relevant faculty and administrators.

III.1 Evaluation criteria

A. Instructional activities

Unlike the faculty cumulative performance in scholarly activities, which can be systematically documented over the years and linked to the effort of the individual faculty member, the effectiveness of the instructional activities of a faculty member, except in highly unusual circumstance, can not be measured by the success or failure of an individual student (or a group of students), who might have been impacted by this faculty member in the past. To link directly the effort of an individual faculty and his/her impact to an individual student would require tremendous resources to follow the future careers of all students (or, at least, a sufficient number of students even just for statistical studies) after their graduation (also, perhaps, those students who fail to graduate). Given this and a few other difficulties, the teaching evaluation is often limited to the student evaluation, and sometime only to a small number of numerical scores compiled from a student survey of the (short term) classroom experience administered before the end of each semester.

  1. Student Evaluation - The current student evaluation at USC was derived after considerable faculty and student input. To centralize the process, most units utilize a standard evaluation template prepared by the Evaluation Services that provides numerical scoring and allows narrative input in response to standard questions. We believe that the current evaluation form does a reasonable job in measuring the students' view of the quality of instruction, especially when statistical information is accompanied by the narrative comments, which often yield key insights. It should be used in every undergraduate and graduate course taught at USC. However, we also recognize a possible flaw in the process, namely, that it does not necessarily distinguish between the popularity of the instructor and the quality of the learning experience. In addition, it is a common experience that, for administrative purposes, only two of the twelve questions of the current evaluation are used in a typical teaching evaluation, specifically (#11) "Overall how would you rate this instructor" and (#12) "Overall how would you rate this class". Thus, much of the standard form is underutilized.
  2. Other parameters - The distribution of teaching loads should be equitable, but the specific form of "course equivalents" may vary not only from discipline to discipline, but even between subjects within a department. The "quantitative" parameters of a faculty member's teaching load may include:
    • lecture hours per week,
    • class size and course management,
    • nature of the course (elective, required, general education, graduate, undergraduate, etc.)
    • degree of difficulty in course preparation and grading,
    • laboratory or discussion sections, and
    • course development (e.g., new course or revised course),
    Two other "qualitative" parameters, could also help in the teaching evaluation, namely,
    • post-graduate evaluation (e.g., exit survey at graduation) and
    • peer evaluation including review of syllabi and, perhaps, actual attendance at lectures.

B. Scholarly activities

Most faculty members are familar with the peer evaluation based on the impact and the quality of the works of their peers. They may not be as critical in the evaluation of their colleagues in the same department, but, we believe that they are sufficiently experienced to differentiate the subtle differences in the quality of scholarly works.

C. Service

It is generally agreed that the evaluation of the service activities should be left to the Chair/Dean and the individual faculty member. External acknowledgements (e.g., letters from the Committee Chairs, Faculty Council, Academic Senate, Deans and other Senior Administrators) should be included in the consideration.

III.2 Evaluation process - an example

To illustrate the merit evaluation process at USC, as an example, we briefly outline in the following the current procedure employed in the College of Letters, Arts, and Sciences. (A more detailed description is given in the Appendix.) Specifically, it consists of the following steps:

  1. Submission of Annual Activities Report by faculty member,
  2. Election of the Departmental Merit Review Committee and evaluation by the Merit Review Committee,
  3. Review of the Faculty Evaluation Report by the Department Chair,
  4. Response by the faculty member,
  5. Submission of completed reviews to the Office of the Dean of Faculty, and
  6. Appeal of merit evaluation.

IV. Salary Recommendation

Again, we use the procedure employed by the College of Letters, Arts, and Sciences at USC as an example to illustrate the procedures that link the merit evaluation to the salary recommendations. At the conclusion of the merit evaluation, assuming that the budget for the following year is already (or will be soon) established by the Provost's office, the Deans of the school will consult with the Chair concerning the salaries proposed by the Chair based on the merit evaluation, to determine the salary recommendation for each faculty. Any faculty member who believes that the salary established for the next year is inappropriate may appeal to the Deans through the Chair. Appeals beyond this level are made through established faculty grievance procedures.

Clearly, the formula used by the Chair to determine the salary proposal is critical to the credibility of the entire merit-based faculty compensation system. We strongly recommend an open communication between the Chair and the faculty on the guidelines for the salary recommendations. It is also the responsibility of the Deans to encourage, or to ensure if possible, such a communication.

At the school level, a special salary pool may be established for retention, promotion, incentive for faculty rejuvenation, and adjustments associated with other long term problems such as the salary compression and gender/ethnic inequities, if any. It is the responsibility of the Deans and the Chairs to identify potential salary problems before serious negative impacts materialize.

Another key issue in the development of a merit-based faculty compensation system is whether the salary recommendations within the same academic unit can accurately reflect the outcome of the merit evaluation, based on the same criteria and judged by the same peer group. There will always be a certain degree of faculty dissatisfaction due to misunderstandings as long as the salary structure requires confidentiality. At the very least, a publication of the salary "quartiles" by rank and years in the rank at the school level may help faculty to better determine if the recommended salary reflects accurately the cumulative performance as judged by his/her peers. It may also help the faculty to identify other long term problems including salary compression and other inequities, if any.

To maintain the confidentiality, the faculty salary increase is often measured quantitatively by a percentage increase, both in the determination of the annual salary pool and the individual salary increase. A higher-than-average percentage salary increase for a junior faculty with lower base salary may mean a less-than-average increase in dollars in comparison with the ones enjoyed by a senior faculty with a substantially higher base salary. On the other hand, for a senior faculty, with an outstanding cumulative performance rating (even though his/her short term "productivity" may be less than outstanding), it is difficult to accept a "less-than-average" percentage salary increase, in spite of the fact that his/her increase is actually higher-than-average in dollars. In practice, of course, in the actual determination of the salary increase, the calculation is often carried out initially in dollars. A shift in quantitative measurement from a percentage increase to an actual dollar increase, or, a combined use of both, may remove some of the emotional factors associated with a compensation system which may appear to be less than equitable by many.

V. Other Issues

Finally, we would like to include in our discussion two additional issues, which may be beyond the scope of the establishment of a faculty compensation system, but they are intimately associated with our activities as a faculty member.

First, excellent research represents a serious time commitment, whether it requires space and test tubes or books and a word processor. If teaching loads remain high in certain areas, they seriously restrict the time available for first rate research in those areas. The instructional profile of the individual faculty member may be established jointly by a group made up of the faculty member and the given department's personnel committee and/or chair, and approved by the Deans. The process of establishing the profile should include a detailed review of the faculty member's past and current activities. Assigning equitable teaching loads, providing equitable amounts of time for research, and fostering a sense of fairness across the school will raise the research profile of the school as a whole. Serious consideration should be given to the appointment of additional faculty if necessary.

Second, increased interaction between faculty and students is both desirable and necessary in order to improve student retention rate. The advent of information technology offers a tremendous opportunity for breaking down the initial barrier between students and faculty. Unfortunantely, much of this opportunity is being lost because information about faculty is not widely available to students electronically and most faculty do not have easy access to the type of student data that is critical for effective student advisement and mentoring. A better understanding by the students of their faculty through a well designed faculty data base linked directly to the home page of each faculty member, which summarizes his/her instructional and scholarly activities, course syllabi (or, summary of past student evaluation), and other information (of course, with individual privacy strictly protected), could become one of the most important steps in bringing together the students and faculty at USC. The faculty data base could also facilitate collaborations between faculty once they become better informed of the activities and interests of their colleagues.