|
What is enrollment management? Here is a hint: If you work for an institution of higher education, then your institution is already engaged in enrollment management. There are, however, two ways to proceed: strategically and non-strategically. In an institution without a consistent approach to enrollment management, the different components of enrollment management are likely to be working inefficiently or even at cross-purposes, yielding poor results. Many aspects and processes of higher education comprise enrollment management—planning, marketing, placement, research, scholarships, retention programs, student services, scheduling of classes…to name a few—but unless efforts are coordinated to achieve an institution-wide effect, the process is not truly strategic.
Strategic Enrollment Management (SEM) is not necessarily simple or easy. Dolence (1996), describes SEM as follows:
Strategic Enrollment Management is a comprehensive process designed to help an institution achieve and maintain the optimum recruitment, retention and graduation rates of students, where optimum is defined within the academic context of the institution. As such, SEM is an institution-wide process that embraces virtually every aspect of an institution’s function and culture.
Most authors and practitioners agree that SEM requires a commitment from the top institutional leaders to encourage and to fund initiatives. In California, respecting shared governance processes is also necessary to create “buy in” and consensus about strategies for defining and realizing optimal growth. The stakes for successful implementation of strategies are high. As Penn (1999) puts it, “Over the years, colleges and universities have closed, in part because they did not pay attention to their revenue stream or the impact of enrollments. Campuses have also begun to realize that many different programs, services, and structures affect enrollment, yet these different influences are not coordinated. In turn, fluctuations in enrollment shape institutional processes and planning. New curricula may need to be developed because of changes in enrollment, housing or parking options may need to be changed based on the age of students, and program development may need to be altered by lower retention rates.”
Enrollment management is about money and access; success and failure; supply and demand; monitoring growth and decline; and anticipating change. In this article I will explore three core aspects of enrollment management that an institutional researcher is well-positioned to address: segmentation; enrollment forecasting; and marketing. I will flesh out the discussion of these general techniques with some examples based on my experiences as Director of Institutional Research of Mendocino-Lake Community College District, a rural district based in Ukiah, California, about 120 miles north of San Francisco.
Segmentation
Community colleges and California community colleges in particular, have a broad, mandated mission. There are many different constituencies that avail themselves of the opportunities a local community college provides. Some common distinctions that are drawn include traditional students, returning or non-traditional students, vocational students and lifelong learners (a.k.a. recreational or avocational students). It is helpful to go beyond these categories and examine enrollment patterns by age, geography, ethnicity, gender, educational goal and unit load. These well-defined variables create segments of the student body that are relatively distinct from other segments.
The segments that come into play at any particular campus will depend on the characteristics of the local population and the college’s offerings. They will become the lenses through which the leadership is able to monitor trends in enrollment and create plans to optimize growth (or minimize cuts). At Mendocino College, for example, segmentation of the student body revealed that there are three main age groups that carry different average unit loads. Those students who were 40 or older represented 55% of the college’s headcount, but only 35% of the college’s enrollment, whereas those students of more traditional college age (24 and younger) represented just 35% of the college’s headcount, but 55% of the college’s enrollment. The younger students carried, on average, a much heavier unit load of classes. Those student in the middle age range of 25 to 39 are by far the smallest group and their unit load was somewhere between the heavy load of the younger students and the light unit load of the older students.
Segmentation, while interesting in its own right for what it tells us about our community and our students really comes into play when it is used to hone predictions about student enrollment in upcoming semesters. Different people respond to the same situations differently according to their needs, abilities and goals. In a very volatile fee environment such as we have experienced in 2003 and 2004, the different segments of a college’s population can expected to have their own, sometimes opposite reactions.
Segments worth analyzing:
- Age group
- Gender
- Ethnicity
- Concurrently enrolled students
- Full-Time versus Part Time
- Vocational versus non-vocational
- New versus Continuing
- BA holders
Enrollment Forecasting
The importance of enrollment forecasting (and SEM) has to do with the community college’s mission to provide access. It is not possible to provide access, however, if appropriate funding is not secured. Thus, SEM provides a method for assuring a steady stream of funding for the mission critical activities of community colleges—such as enrolling and educating students for college credit and vocational education.
It is necessary to understand that funding for California Community Colleges comes not from enrollment fees but from taxpayer money that has been funneled through Sacramento. State funding formulas dictate that community colleges receive funding principally based on the number of Full Time Equivalent Students (FTES) that they serve (there are other aspects and some intricacies, but in the main, FTES rules the day). Growth funding from the state—in the current year at least 3% growth over the previous year will be funded—creates a perfect target for defining what “optimal” means to a district. In an environment of increasing fees, however, enrollment growth is far from assured for all colleges. Increasing enrollment fees has been seen by some as a back-door method to regulate enrollment in community colleges and thereby reduce the funding to community colleges by limiting or reversing growth (Community College League of California, 2003; Perry, 2003).
In 1993, McIntyre and Beno spearheaded a collaborative analysis of the effect of three major fee hikes in the California Community College system. Their research revealed that the laws of supply and demand do in fact apply to the “purchase and consumption” of education. In spring 1993 community college enrollment fees increased from $6 per unit to $10 per unit. Additionally, students who already held a Bachelor’s degree were required to pay a “differential fee” of $50 per unit. The general enrollment fee increased again in fall 1993 from $10 to $13 per unit. The 1993 report found that enrollment was reduced by 41% among BA holders and by 4% for the rest of the community college population in spring 1993. By fall, colleges had responded to reduced student demand by reducing the number of class sections available, also there was another fee increase of $3; this combination caused further enrollment reductions of 2% from spring 1993. The effect of these fee hikes indicated that without a doubt that economic factors are powerful predictors of enrollment.
Further confirmation that college enrollments are price sensitive (i.e., that demand is elastic) comes from a research study in which Kane (1999) compiled data from 3,000 schools nationwide and concluded that at two year colleges enrollment fell 14.9% for every $1000 increase in annual cost (versus 10% at four year schools). A more segmented approach taken by McPherson & Shapiro (1998) found a similar level of demand elasticity among low income students: an increase of $150 reduced enrollment by approximately 2%. Clearly these numbers are significant.
Data on past trends is typically used as a basis for predicting the future. The California community college economy has, as a whole, reacted to a financial crisis approximately once every ten years. The extent to which the crisis is realized in each independent district and campus depends on the nature of the local population as well as the institution’s foresight, preparation and reaction to the external events of recession and reduced state funding.
A General Approach to Predicting Enrollments
Let the past be your guide; you can use previous enrollment trends to create models of the effects of different external and internal factors. Begin by dividing your influences into those that increase enrollment (drivers) and those that decrease enrollment (sappers). This will aid in the discussion of strategic initiatives. Strategies can be devised to counter a negative (e.g., high school graduates are declining, therefore we need to increase our share of those who remain) or to accentuate a positive (e.g., our biology programs are constantly full, we can add a new full-time faculty member and increase our offerings). Some of these factors (e.g., fee increases) will be able to be modeled with a fair degree of precision, thanks to existing data and prior research. Other factors (e.g., new retention initiatives, previously unmodeled influences or complex influences, such as policy changes at the UC and CSU) may have to be factored in as best they can. This uncertainty is appropriate in forecasting and can be used to create upper and lower limits on likely outcomes.
Because models include assumptions, scenario planning is a useful technique to combine with enrollment forecasting. Scenario planning involves varying the parameters of certain assumptions. One parameter, for example, is change in enrollment fees. One model might assume that the legislature will again increase enrollment fees next year. Will the BA differential fee return from the abyss? What effect would such a fee have on enrollment?
Scenario planning creates different “what if” situations. Typically three scenarios are created: a best-case scenario, where all the factors fall in as best as could be expected; a worst-case scenario; and a middle of the road, “most likely” scenario (more complex situations may require more scenarios, but more than five scenarios may difficult for most people to digest). Different contingency plans can be prepared for each scenario. In this situation, it is important that projections are clearly differentiated from enrollment goals. A district may have the goal of growing 3% every year, but the projections may indicate that 1% growth is much more likely. If the president or chancellor is serious about growing 3%, then initiatives need to be started and funding will have to be allocated to pursue the desired, but not expected, growth.
At Mendocino College, the basic econometric model led me to expect that increases in fees would result disproportionately in a decline in the enrollments of older students in fall 2004. This decline would be due to a greater demand elasticity of older students who are often not eligible for financial aid and fee waivers as they reacted to the higher enrollment fees. Essentially, I started my enrollment projection with a two segment approach to forecasting the enrollment for 2004-2005. Students who were eligible for Board of Governor fee waivers were considered to have an inelastic demand, because financial aid would pick up the additional expenses. Enrollments by students who were not eligible for fee waivers, however, were predicted to decline 3.8%, based on previously demonstrated levels of demand elasticity from prior fee increases.
Demand elasticity is likely to vary district by district. Whereas previous research had found an elasticity of -0.7 for all California Community College enrollments, I found that non-BOGG fee waiver students at Mendocino College had displayed and elasticity of -0.4, showing that local variations in elasticity can be quite significant. Further details on econometric modeling can be found in McIntyre & Beno (1993) at: http://www.cccco.edu/reports/attachments/1993_fee_impact.htm
After establishing a demand model of enrollment, I factored in several other key components: size of the graduating classes at local feeder high schools (available at http://data1.cde.ca.gov/dataquest); the expected results of a new retention program; increases in enrollment due to policy and fee changes at CSU and UC; and an expected increase in student athlete enrollment. By varying the factors from their lowest to their highest and to their most likely, three different scenarios were produced for consideration by the president and the rest of the campus community.
Seeing projections of enrollment decline can have a galvanizing effect, if handled appropriately. The demonstrated need for a retention program to turn in a good performance, for instance, can create the will and resources to ensure that such an outcome occurs. The dynamic nature of higher education enrollment means that as prices increase at the UC and CSU, the Community College system becomes, relatively, a greater bargain. Still, the higher fees at the community college will drive away some students, particularly older students who are not eligible for financial aid. It is important to note that projections become increasingly inaccurate the further out they go. Most projections do not have much usefulness beyond a five year horizon due to the inherent volatility of the assumptions used to create the projection models.
Key factors to include in an enrollment projection:
- Fee increases and elasticity of demand, preferably segmented by fee waiver students vs. non-fee waiver students
- Size of graduating classes of local high school feeder districts
- Capture rate or yield to the CC of local graduating classes
- The likely effects of other programmatic factors and initiatives
- Population increases, weighted by segment
- Unemployment rate and/or other economic proxy data
- Create different projections for different segments; sum the results for a total projection
Marketing
A fancy report that uses multiple regression to forecast enrollments is all well and good, but how do you communicate your results and to whom? The internal audience is as crucial as the external audience. “Responsibility for enrollment management begins with the president and extends throughout the institution, with the participation of faculty being of critical importance” (Mabry, 1987). Deans and chairpersons of academic units are also in a key position to participate and influence enrollment management. The internal audience will largely dictate the degree and form of the institution’s response to enrollment projections and other enrollment management information.
Communicating with the external audience—the community—is a task that largely falls to the marketing department. It is also an integral aspect of enrollment management (though some authors consider it a distinct area of admissions management). It is through marketing that students are contacted and informed of the opportunities that await them at their local community college. The college can shape its student body by tailoring its marketing campaigns to specific market segments that have been identified as congruent with the college mission.
Research is an integral part of marketing and the institutional research department should maintain a close working relationship with the marketing department, providing it with more than fodder for news releases by involving marketing in discussions about SEM. Targeting specific segments is one area that is appropriate for collaboration between marketing and research.
At Mendocino College, I collaborated with the Director of Public Information to create and execute a marketing survey of our student body. Using the data culled from this survey, marketing is now able to target specific advertising messages to key demographic groups. For instance, segmentation and trend analysis revealed that a traditionally large component of enrollment, women over 40, was fading and enrollments were dwindling. Now, marketing is able advertise with the specific local publications and radio stations that are favored by this demographic group and deliver a message that appeals to their specific concerns and traditional areas of interest.
Conclusion
In this article I largely focused on SEM as a tool to increase enrollment. Management of enrollment can, of course, include the need to regulate or restrict growth and this is sometimes necessary. Reduction of offerings is the most direct way to achieve this and enrollment management in connection with the strategic guidance of the mission and college planning efforts can direct the closure of sections to areas that create the least disruption and maintain a balance of offerings. Other initiatives, such as the use admissions criteria for impacted Nursing programs, have shown that access can actually be enhanced by the use of selection criteria to optimize student retention and success.
Researchers are able to bring many components of SEM to the table, particularly the understanding of the scope and purpose of SEM, but no researcher or enrollment manager will be able to be effective all by themselves. Presidents, vice presidents and deans need to consider the demographic composition of the students they serve; the financing choices students are likely to make; services needed to help students balance work, family and academic obligations; and how much an extra year of college costs students in terms of expenses and lost income. An institutional commitment that begins with the president and includes a number of key staff will be necessary to integrate enrollment management into a truly coordinated and strategic process that is capable of meeting its goals.
References
Community College League of California, Legislative Office (2003). Internet citation: [http://www.ccleague.org/leginfo/]
Dolence, Michael G. (1996). Strategic Enrollment Management: A Primer for Campus Administrators. Washington, D.C.: American Association of Collegiate Registrars and Admissions Officers. ED 384326. 29 pp.
Heller, Donald E. (2001). The Effects of Tuition Prices and Financial Aid on Enrollment in Higher Education: California and the Nation. Rancho Cordova, Ca: EdFund. [http://www.edfund.org/pdfs/i-57.pdf]
Kane, Thomas (1999). The Price of Admission: Rethinking How Americans Pay for College. Washington, D.C.: The Brookings Institution Press.
McIntyre, Chuck & Beno, Barbara (1993). 1993 Report on Fee Impact. California Community College Chancellor’s Office: Sacramento, CA. [http://www.cccco.edu/reports/attachments/1993_fee_impact.htm]
McPherson, Michael S. & Shapiro, Morton O. (1998). The Student Aid Game: Meeting Need and Rewarding Talent in American Higher Education. Princeton: Princeton University Press.
Mabry, Theo N. (1987). Enrollment Management. ERIC Digest. Eric No. ED286558.
Penn, Garlene (1999). Enrollment Management for the 21st Century: Delivering Institutional Goals, Accountability and Fiscal Responsibility. ASHE-ERIC Higher Education Report Volume 26, No. 7. Washington, D.C.: The George Washington University, Graduate School of Education and Human Development.
Perry, Patrick C. (2003). Access Lost: An Examination of Supply Constriction and Rationing in the California Community College System. Sacramento: California Community College Chancellor’s Office.
|
