Groups responsible for driving the NCAA's ongoing review of initial-eligibility standards have been given what amounts to a new set of wheels in the form of academic transcript and graduation data from the 1994-99 Academic Performance Census (APC).

The data, available for the first time, provide key up-to-date information for the NCAA's initial-eligibility consultants and other groups as they steer toward possibly changing the Association's eligibility standards.

The APC is expected to form the backbone of the evaluation of existing NCAA eligibility rules, as well as the potential development of new rules. Due to the large and representative sampling, and the increased reliability of some of the measures used, the current predictions of freshman academic performance from the APC are considerably more accurate than previously reported in NCAA or other academic research.

"It gives groups like ours just what they have been looking for, which is information that will lead to developing standards that better predict success in college -- with ultimate success being graduation," said Charles Lindemenn, former director of athletics at Montana State University-Bozeman and one of the NCAA's initial-eligibility consultants, a group appointed by the Division I Board of Directors two years ago to monitor the health of Prop 16.

Although only a single cohort has been fully reported, the APC results are nonetheless significant. Not only do they confirm findings from the Academic Performance Survey (an instrument used to track academic success before Prop 16), they also show that student-athletes come to school with slightly better preparation, on average, than the general student body, and that high-school variables are much more reliable predictors of early college behavior than they are of college graduation.

In such an approach, standards that predict success in college would eventually rely less on high-school variables and more on in-college performance, but the overall prediction system from high-school to college graduation would be more accurate.

"One use of this information might be to use the continuing-eligibility rules to help identify and then adjust the academic path of student-athletes who are not on track to graduate as measured by those early markers in college," McArdle said.

A change of thinking

While the Division I Academics/Eligibility/Compliance Cabinet is just beginning to delve into whether a seamless approach of blending initial- and continuing-eligibility standards is the best approach, the APC data provide additional empirical evidence for those who are advocating that route.

"What we're finding is that trying to predict only based on high-school work is probably not as good a system as trying to create a seamless model where we have benchmarks that students need to reach, beginning with high school and extending through their college careers," Lindemenn said. "So it is true that high-school grades and test scores better predict success at the end of the freshman year than they do predicting the full extent of what's going to happen throughout college and eventually graduation."

Those findings, said Lindemenn, likely will prompt the initial-eligibility consultants and the AEC Cabinet to search for standards that would identify what kinds of "academic profiles" students need to accomplish at the end of high school and during each year of college in order to be positioned to graduate.

"What that's resulting in is a change of thinking," he said. "We've always talked in terms of continuing eligibility, but now we're talking in terms of 'progress toward graduation.' "

"These and other data indicate that the best single predictor of eventual graduation is the number of quality points achieved by the end of the second year in college," McArdle said. "However, the new data show that freshman GPA and freshman credits earned predict graduation almost as well. Once you know a student's GPA and credits earned in the first year, the high-school variables no longer are needed to optimize the prediction of college success. The first two years in college may represent the most critical turning points in the academic life of a student-athlete."

Lindemenn said the new results may be a wake-up call for a system of academic progress that perhaps is outdated.

"In the past, we set a standard that was so low in terms of continuing eligibility that students who were on the margins were not in position to succeed," he said. "If all you did was major in eligibility, you couldn't succeed. We need to change that standard. We can't talk about continuing eligibility anymore, we have to talk about positioning to graduate."

What's next

The cabinet subcommittees and the initial-eligibility consultants have asked the NCAA research staff to examine some practical models that would demonstrate the effect of the so-called seamless approach. While no one is certain what those models might look like -- whether the sliding scale of initial-eligibility might change or whether high-school variables might be weighted differently -- the groups responsible for the review are poised for action.

"We want to know how radical a change would be required before we go forward with specific rules. Obviously, if we find that there had to be fundamental changes in the ways institutions and students did their work, then that could be problematic. On the other hand, if it seems like changes on the part of institutions and student-athletes would be relatively minor or manageable, then we think this seamless approach would be an excellent direction to go."

The APC data and results also will continually be improved. McArdle said in the next set of analyses, researchers will look more closely at the early college performances of all the 1995 through 1999 cohorts. Those analyses will indicate the degree to which the findings from 1994 are consistent over time.

Researchers also are hoping to further fortify the data from all APC cohorts. About 25 percent of Division I institutions have not yet provided individual student-level data, but still have the opportunity to do so.

"We hope that all NCAA member institutions will eventually become full members of these APC studies in the near future," McArdle said. "The value of these nationally representative data to the Association cannot be overemphasized."

Eligibility models

Components of a 'seamless' eligibility model

Facts

Grades and test scores used in combination have been shown to be valid predictors of academic success in college.

However, research has shown that first- and second-year outcomes can be predicted nearly twice as accurately as graduation using high-school variables.

New NCAA research has shown that high-school academic performances are not predictive of negative academic outcomes in college once current college academic behavior is accounted for.

The "seamless" model is an attempt to improve the ability to predict academic success, and to implement intervention strategies that will lead to higher graduation rates while minimizing disparate impact.

Goals

Attempt to find "markers" of graduation (for example, number of quality points) in first- or second-year academic performance of student-athletes.

Use initial-eligibility model to predict that marker.

Use continuing-eligibility model to intervene if student-athlete appears at risk based on academic markers.

Find appropriate penalties/incentives to achieve desired academic behavior.

Maintain graduation as the ultimate outcome.