Preliminary data from the Academic Performance Census has prompted the Division II Academic Requirements Committee to examine whether legislative changes might be necessary to initial-eligibility requirements in the future.

At its recent meeting, the Division II ARC reviewed results of the first two years of APC data (entering classes of 2006-07 and 2007-08) showing that high school grade-point average may be a much stronger predictor of first-year performance in college than scores from standardized tests.

Similar findings over time led Division I to drop the cut score on tests as part of its sliding scale to determine initial eligibility. Division II requires prospective student-athletes to earn at least a 2.0 GPA in 14 high school core courses and an 820 on the SAT (or 68 on the ACT) to be eligible to compete as freshmen. Data from the first two years of Division II APC collection, though, show that about 7 to 9 percent of prospects are either partial or nonqualifiers.

While Division II ARC members agreed to discuss at future in-person meetings how test scores can be used more effectively in certifying initial eligibility, they also know that the first two years of data isn’t enough upon which to make an informed decision about eligibility standards right now. In addition, the high number of schools using the data-entry system for the first time resulted in a slightly higher rate of data errors than the committee would like. These errors will be corrected in the next round of data collection.

“It’s too early to state definitively that GPA is much more meaningful to Division II eligibility than test scores,” said ARC chair Paul Leidig, the faculty athletics representative at Grand Valley State, “but the data we have at least from this cohort show that very thing.”

The 2006-07 and 2007-08 cohorts are the first collected since Division II began requiring schools to submit academic data as the result of legislation passed at the 2008 Convention. While it may take another year or two before the data are plentiful enough to drive decision-making on eligibility standards, the preliminary results have at least piqued ARC members’ interests.

“It stands to reason that the high school GPA is a good predictor of college eligibility because we tie the eligibility to the GPAs of those students, especially after they get in to college,” said ARC member Kevin Schriver, the faculty athletics representative at Southwest Baptist. “The entrance exams do have validity, but it does not surprise me that you start to get this indicator that high school GPA is more closely associated with college GPA because you’re comparing apples to apples.

“But for right now we’re still looking at how much the high school GPA configuration contributes to that predictably and how much the test score contributes to that predictability.”

The committee directed the NCAA research staff to analyze whether the minimum 2.0 GPA requirement for initial eligibility is too low to predict first-year outcomes. The committee recognized the need to evaluate all of these issues by balancing the potential disparate impact on different segments of the student-athlete population.

“Over the next period – and it has to be more than just a year – we want to be taking a close look at the breakout between grades and scores,” Leidig said. “At this point, high school GPA seems to be a much more accurate predictor of academic success in college. But it’s way too early to discuss whether these data would lead to Division II incorporating some sort of sliding scale similar to what Division I uses.”

The Division II ARC meets again in February. While a decision on whether the data lead to changes in eligibility standards is unknown at this point, the good news for ARC members is that they will have an expanding set of reliable data in the coming years upon which to make more informed decisions.

“If next year’s cohort shows an even more striking difference, then that may expedite our thinking, but if it’s only marginally different, then that’s a different story,” Leidig said. “Right now it’s too early to make the kinds of ongoing observations that a fuller data set would prompt.”