« February 2010 »
1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
You are not logged in. Log in
Entries by Topic
All topics  «
Blog Tools
Edit your Blog
Build a Blog
RSS Feed
View Profile
Navigate Story Archive
Friday, February 19, 2010
Inflating Carolina: Achieving a workable solution

GRADE INFLATION IS SUCH A PROBLEM for the University of North Carolina that I can’t even find a list of UNC's recent valedictorians.  Why not?  Probably because printing it out would cause a nationwide paper shortage.  Earnestly, after decades of grades lofting into the stratosphere, is there anything our university (and others) can do to reverse grade inflation, or at least to compensate for the progressively more meaningless and misleading GPA?

In 2007, when grade inflation last came up for discussion at UNC, some professors and students opposed any system that rations grades, including quotas like those implemented at Princeton University and mandatory departmental averages like at Wellesley College.

“My old boss called these the power tools.  It’s amputation for a small injury,” said UNC professor Andrew Perrin, one of those who studied grade inflation for the faculty. “But it does have the advantage that it is very transparent, and it’s easy to explain to the outside community," Perrin added.

Happily, there exists a better, if more complex, system for ranking students.  A proposal made to UNC several years ago is back: add the Achievement Index to students’ grade reports, a measure that filters out variations in grading among different academic departments and among individual professors.  It is a great idea.

As the Daily Tar Heel describes it, the Achievement Index is a “strength of schedule” analysis.  The Achievement Index (AI) measures a student's performance against their classmates’ performance given those classmates' grades in other courses.

The Achievement Index was developed by a biostatistician, Valen Johnson, as a method for combining the information from grades earned across college courses; where the overarching goal is to measure each student’s academic performance while factoring out differences among individual instructors' grading practices.  (Further information on grade inflation is available in the book "Grade Inflation: A Crisis in College Education" by Valen E. Johnson, and a briefer synopsis of the technical aspects of the AI can be found by Googling "Primer on the Achievement Index".)

The basic concept is this: an A in a class where everyone got an A tells us nothing.  An A in a class filled with classmates whose grades in other courses were poor is less of an accomplishment than a B in a course filled with classmates with A's in other tougher courses.  Put differently, an A in a crip course is not as telling as a B in a killer -- as every college sophomore knows.

And research shows the AI works better than the GPA -- even using the exact grades now used to calculate the GPA.  In 2007, UNC’s faculty Educational Policy Committee studied the validity of the AI with two tests comparing pairs of UNC students enrolled in the same class at the same time. GPA and AI were assessed by examining whether GPA ranking or AI ranking better predicts these students’ performance in the very same class.  In 61% of 22,000 cases where students in such pairs earned different grades in the same class, the higher grade was earned by the student who ranked higher on AI but lower on GPA.  

Likewise, based on 14,000 pairs of students in exactly the same class, where one student would have received a degree with distinction using AI (but not GPA), the AI-honor student earned a higher grade than the student with a GPA-honor almost twice as often.  So, clearly, for both assessments, the AI is a better, more valid measure of academic performance.

In the spring of 2007, UNC's Educational Policy Committee recommended that Carolina list their AI on students’ transcripts along with their GPA, that UNC award graduation honors based on AI and also use AI to determine class rank.   After what was described as "a spirited debate," the faculty governing body voted 34-31 against adopting the Achievement Index.   We can hope, in the meantime, at least two faculty members have gained some common sense.

Gary D. Gaddy’s nephew, Benjamin Gaddy, was one of 129 valedictorians at North Carolina State University in 2007.  He had a 4.0 in electrical engineering.  Who do you think should have been THE valedictorian?

A version of this column was published in the Chapel Hill Herald on Friday February 19, 2010.

Copyright  2010  Gary D. Gaddy


Authored by Gary G. Gaddy at 8:01 AM EST
Updated: Friday, November 5, 2010 10:01 AM EDT
Post Comment | Permalink

View Latest Entries