One of the themes in the much commented on report of the Spellings Commission highlights the need to fully inform higher education consumers about everything. For some, accountability not only means being responsible about teaching and research, but also delivering some form of full disclosure. This trend reflects the continued move of higher education from a specialized product sold to well-informed customers to a generic product sold in widely varying formats to large numbers of often unsophisticated consumers.
As is usually the case with high profile commissions, this one responds to a mature trend, not something new and different. The proliferation of rankings and ratings of every conceivable type is the clearer example of the commodity college degree, but the commission, because it speaks for at least one part of the government, has a coercive capacity where the ratings have only a demonstrative capacity.
What, then, is the full consumer information we need? Much current university and college published data is actually not very helpful. As a normal practice, we produce measures of central tendency -- averages or means -- or we provide ratios of one kind or another. So we talk about average class size or average student/faculty ratios; average discount rate on tuition and fees; and the average financial aid package or the average debt on graduation. Universities and colleges provide information on the average endowment or average state investment per student.
All of these, and many others, provide an average representation of the reality of campus life. If universities and colleges managed, as do other high tech, high quality enterprises, by reducing the variation around the mean to produce a homogeneous product, these average numbers might have some usefulness. That’s not how higher education works.
Instead, colleges and especially large public universities manage in ways that appear to maximize the variation they can sustain in the quality and diversity of their students. They admit students with SAT scores ranging from 900 to 1600 perhaps, students whose parents have no taxable income and those whose income reaches above six or seven figures. They admit students who are the fourth generation of college attendees and the children of migrant workers whose home experience includes no prior engagement with higher education. Universities pride themselves on the wide diversity in the ethnicity and economic capability of their students and they speak eloquently of the wide range of socioeconomic circumstance from which their students come.
This is all to the good, but it illustrates why the average numbers we often discuss as the tokens of accountability disguise more often than they inform. Instead of average class size, we might display the percentage of students in classes under 25, 26 to 50, 51 to 100, and over 100. Even that is not as helpful, for example, as providing a transcript analysis of the graduating class. The aggregate measures that tell us how many classes are under 50 students tells us how the faculty teach, but not what individual students take. Students in engineering may have mostly classes smaller than 50 while students in humanities or social sciences may have mostly classes larger than 100. We may find that 30 percent of our graduating students never took a class under 50 even though such classes were available. Knowing what kinds of class contexts are available is a helpful overall indicator, but it does not tell the interested consumer what students actually choose to do or are advised to do.
We call for better information on the cost of college. By this, we mean both the “costs” of what colleges spend on providing an education and the “price” that students pay for that education. The latter is a very slippery number. Everyone knows that there is a sticker price and a discounted price. Everyone knows that students receive discounts for various reasons.
What we do not provide very often are data that describe the characteristics of students who receive discounts and reveal the relationship between particular characteristics and the discounts the institution provides. For example, we do not know the relationship between the marker for merit (SAT, GPA) and the amount of merit aid provided (for those institutions that provide merit aid). If we did, we might find that not all students with a 1350 SAT will get the same merit aid package.
Almost all institutions provide a wide range of need based aid, some from federal or state sources that are regulated and some from institutional sources that are not. Institutions create need based packages to achieve enrollment goals, and sometimes following a formula based on the federal guidelines and sometimes using ad hoc packaging to achieve balance in our student populations. This is especially so when institutions are under clear directions from their boards to change the composition of the student body in some way, for example to prefer legacies or first generation students, or to increase the percentage of men or women.
Student debt is a mystery number because the data on average debt deal with only a fraction of the student population. Average debt refers to the average institutionally managed debt of those graduating seniors who have debt. So it does not tell us about the debt of those students who in addition to institutionally managed debt have private debt from a local bank, from credit cards, or from other sources. It also does not tell us about those students who do not qualify for any institutionally managed loans but nonetheless borrow money from local banks, credit cards, and other sources. Nor does it tell us how much of the debt students contract is required by the formal cost of attendance and how much responds to lifestyle issues related to housing, transportation, illness, family obligations, and entertainment among other issues.
In the real world of higher education -- rather than the idealized world of commissions and homogenizing government regulations -- higher education institutions, while they produce a standardized product, do so for widely varying market niches made up of customers with widely varying characteristics.
Many of the proposed measures that we see coming from commissions and regulators speak to some mythical average student experience, usually reflecting the idealized type of the elite private four-year college. As such they may satisfy some, but will surely fail to provide more accurate information to individual consumers. How, we might ask, am I to know whether my child is average and therefore likely to have the average experience the data highlight? How many of the graduates actually participated in the average experience, or did most of them pass through the institution at the upper or lower edges of the experience represented by the calculated average?
Most university people also know that any significantly useful data will be used against them. If the data are specific and clear, and if they demonstrate differences among universities and colleges, bitter experience teaches that the regulators will praise those whose indicators are high and condemn those whose indicators are low, without paying the slightest attention to purpose, organization, circumstance, or mission of the institutions involved.
In our world, “high” means most like a private elite liberal arts college. “Low” means most unlike these colleges. The regulators, whether accreditation agencies, higher education commissions, public boards of trustees, state regulators, or national agencies, will acknowledge that the institutions differ dramatically, but they will then proceed to compare unlike institutions in statistical grids using the already not very good measures.
Worse yet, the commercial ranking services, once delivered this treasure trove of numbers that bear the imprimatur of regulatory agencies, will construct increasingly convoluted and methodologically flawed rankings, cloaked with regulatory authority, conveniently forgetting the caveats about the data contained in obscure footnotes in the government reports.
Almost every university and college works to get better, driven by the inexorable pressure of competition for faculty, students, funding and prestige. The market place in higher education is exceptionally wide and contains every imaginable modality. The irony of these periodic bursts of regulatory enthusiasm is that the conservative, free-market political enterprise appears most eager to see this educational free market socialized into government mandated homogeneity, often based on simple, one dimensional measures.
Colleges and universities can respond either by resisting all notions of measurement or, more appropriately, by developing and publicizing the detailed, data based measurements that will actually give their many constituencies a clear guide to the complexity and range of the institution’s activities. This is harder than fighting over simple-minded measures, but it surely the right thing to do.