In Affirmative Action, It's Curriculum That Counts
By Clifford Adelman
U.S. Department of Education

So what have law-makers and educators offered lately to "fix" the affirmative action problem at the 200 four-year colleges (out of 1,800) that exercise any degree of selectivity in undergraduate admissions?¹ Lotteries? Admit the top 10 percent of every high school graduating class? Abolish tests? All of these have either been proposed or legislated. However well-intentioned, these quick fixes do little to increase the odds of success for the mass of minority students. It is time to think outside-the-box and come up with more than cheap and easy solutions.

What is the bottom line of "success"? Freshman year grades? Persistence to the sophomore year?² These criteria, traditionally used to develop college admissions formulas, are rather low. Most students attend four-year colleges with the goal of earning a degree, and minority students are no exception. So degree completion must be the criterion against which selection measures are judged. And no matter what kind of statistical gymnastics one performs, the strongest pre-college predictor of degree completion is the academic intensity and quality of one's high school curriculum. Not test scores. As for high school grade point average and class rank, the College Board 1998 survey of high schools reveals that more than half include non-calculations and nearly 20 percent do not compute class rank at all. No wonder the correlation of GPA/class rank with degree completion is even weaker than that of test scores.

It is precisely to GPAs and class rank, however, that legislatures and higher education governing boards turn for the "solution" to affirmative action. I call this the strategy of the Dead White European Male High Command who brought you the first and second battles of the Marne in World War I: if only 100,000 out of a million survived the first battle, we throw two million over the top at the second, and think we have done better when 200,000 come back. There will be a lot of dead bodies out there with the "top X percent" or lottery admissions formulas, and, as usual, a majority of them will be those of minority students.

The best national data--the longitudinal studies of the National Center for Education Statistics--tell a very clear story on this issue. Let us demonstrate with the most dramatic case: by age 30, the bachelor's degree completion rate for all African-American "on-time" graduates of the high school class of 1982³ who entered four-year colleges directly from high school was 42 percent--some 30 percent below the completion rate for the comparable group of white students.

If we restrict the African-American population to the top 40 percent by grades in high school academic courses, the graduation rate increases to 59 percent. Not bad! If we take a different strategy, and restrict the same population to those in the top 40 percent of scores on a general learned abilities test (something like the SAT or ACT⁴), it rises to 66 percent. Better, but not as good as it can get. For if we restrict the population to students who were in the top 40 percent of an academic curriculum intensity scale⁵ and got beyond Algebra 2 to Trigonometry or a higher level of mathematics, 72 percent of the African-American students graduated, and the degree completion gap (compared with a matching group of white students) was cut more than half--to 12 percent.

Which of these indicators would you rather ride in admission decisions? Which does more for students? And which of these is most subject to active, hands-on change? Curriculum. Hands down! If you use all three achievement indicators together, you don't get much further predicting degree completion than curriculum used alone.⁶ This is all common sense, but nobody wants to acknowledge it. Curriculum is the only major pre-college variable you never hear anything about in all the posturing on all sides of the affirmative action argument. Grades are a crap-shoot; a test is a snapshot of performance on a Saturday morning. But a curriculum is an investment of years and provides momentum into higher education and beyond. The effects of grades and tests diminish in time, but the stuff of learning does not go away.

How do we take advantage of the power of curriculum to (1) maintain minority enrollment in selective public universities, (2) increase minority degree completion rates everywhere, and (3) accomplish both ends within the legal restrictions of Proposition 209-type policies and Hopwood-type court decisions? I have three connected suggestions, all of which require colleges and universities to work harder, differently and under more difficult timetables on behalf of their oft-stated goals for minority student representation, and to recognize that the underlying principle is that of equity-not the washed-out, numbing euphemism of "diversity."

1) Change the admissions cycle to one of rolling selection that admits 11 percent of the target pool each month from November through July. Yes, July! Disadvantaged students, among whom minorities are over-represented, benefit more from this schedule. Why? The extra time will keep kids serious through the senior spring of high school, will allow the learning of special outreach and summer programs time to take hold, and will give minority students more time to prepare for and take their tests as late as April or May. Sound outrageous? Not if we acknowledge that the educational systems through which most disadvantaged students pass have slowed them down. Changing the cycle recaptures some of this time and basically adds more than a semester of curriculum to the student's portfolio. It restores some equity in the preparation of minority students for admission to selective institutions. It is an opportunity-to-learn issue. It also would provide opportunity to advise and recruit.

2) Give applicants greater latitude in what tests they present and when they take those tests. Test scores follow a student's curricular experience, not the other way around. And for students who do not test well in restricted-response formats, for example, provide the option of the Advanced Placement exams, which include essay and problem-solving sections, and do not make decisions until those test results are in. More and more minority students are taking AP courses, but AP exams are not used in admissions decisions. It's about time they were! If that means July admissions for September matriculation, so be it! Or let students substitute subject matter achievement tests for the SAT or ACT. At least the subject-matter exams are tied directly to curriculum; and if disadvantaged students are putting more curriculum in their portfolios, they will perform better.

3) Instead of running summer bridge programs for under-prepared students between high school graduation/college acceptance and matriculation, move those programs back at least one year (if not two years), expand them for students from all school districts with historically low rates of participation in higher education, and focus them on curriculum content-nothing else. Total immersion for 10 weeks. Revisit Algebra 2 and move students quickly on to trigonometry and statistics. Pick a real-world problem around which you can build a chemistry or physics curriculum and that requires both laboratory and computer work. Borrow from AP strategies and materials, and have students stitch together a state or local chronicle from original documents and artifacts. If even half of our 1,800 four-year colleges did this for 100+ students from targeted high schools each junior summer, and followed up by establishing Community Technology Centers near each of those high schools where kids could work at terminals on academic assignments and server-fed explorations during non-school time, there will be no second battle of the Marne. These "early bridges," with their emphasis on the academic uses of non-school time, are the only pieces of the package that require real money, but if we are serious about equity, we must acknowledge that it doesn't come free.

All of these strategies, together, will enlarge, if modestly, the academic resources that disadvantaged (and principally minority) students will bring to the table of higher education, no matter where they choose to attend, and will increase their chances of admission at more selective- institutions in their own right.. More importantly, the higher the concentration of their academic resources, the greater the odds that these students will complete degrees. Isn't that what counts for them? For everyone?

* A different version of this article appeared in the Chronicle of Higher Education, Sept. 4, 1998, titled, "To Help Minority Students, Raise Their Graduation Rates."

Endnotes

1 The "affirmative action problem" actually exists only at selective public institutions, i.e. the majority (but hardly all) of the flagship state universities. Reality check: at the most, these institutions account for one out of every six seats in a national entering freshmen class. Selective private 4-year college5, on the other band, sit outside the existing legal framework of affirmative action at the undergraduate level , and can--and will-continue to use race as an affirmative factor in undergraduate admissions. So, for example, affirmative action may now be restricted at UCLA, but it is not restricted at USC.

2 Temporal persistence is a very deceiving indicator. In the national data used for this analysis, 17 percent of those who entered college directly from high school and "persisted" to "year 2" got there with less than 20 credits, and another 12 percent arrived with more than 20 credits but with three or more remedial courses in their portfolios. In other words, three out of ten "persisters" evidenced less than what one would call "sophomore standing." For further details, see Adelman, C., "Academic Resources: Developing an Alternative Index of Individual Student Capital," paper presented to the 1998 Annual Form of the Association for Institutional Research.

3 Data for the analysis in this article come from the restricted files of the High School & Beyond/Sophomore Cohort longitudinal study of the National Center for Education Statistics, specifically from the CD-ROM that includes the high school and college transcripts for this cohort (NCES 98-135). The cohort was established in l980, when the students were in the 10th grade, and was followed by surveys to 1992. The college transcripts allow us to track educational histories to 1993. The author of this article was editor and co-author (with Nabeel Alsalam) of the CD.

4 In all of the longitudinal studies of the National Center for Education Statistics, students are given what can be described as an "enhanced mini-SAT" that correlates at levels above .9 with SAT and ACT test scores.

5 Given the variations in state requirements, academic calendars, and credit system, a national scale of "academic intensity" of high school curriculum was derived empirically from the high school transcripts of the High School & Beyond/Sophomore cohort. At its most intense level, the scale requires students to present 3.75, or more Carnegie units of English -- with no remedial courses, 3.75 or more units of mathematics, with the highest level studied at trigonometry or higher -- and with no remedial mathematics, two or more units of core laboratory science (biology, chemistry, physics), two or more units of foreign language, two or more units of history/social sciences, and more than one Advanced Placement course. There are 30 equal gradations in the academic intensity scale. Once student records are mapped in terms of these gradations, they are analyzed in quintiles (see Adelman, C., "Academic ResourcesÙ," above).

6 In a standard regression model controlling for socioeconomic status, academic curriculum intensity explains about 30 percent of the variance in long-term bachelor's degree completion rates. When test scores and high school GPA/class rank are added to the model, the explanatory factor rises to .341. The odds of any of these relationships occurring by chance are less than 1 in 10,000.

For information about this piece or works cited, contact Clifford Adelman, Senior Research Analyst at the U.S. Department of Education: clifford_adelman@ed.gov