April 18, 2005

An exit exam puzzle

Retired USC professor Lorin Anderson wonders about the difficulty of SC's new high school exit exam:

A retired University of South Carolina education professor who specializes in education trends questions whether the minimum score on the exit exam is artificially low. Lorin Anderson points to a 76 percent pass rate in 2004, while the highest percentage of students who passed a now-retired test on the first try was 70.6 percent in 1991. During the 17-year span the old exam was administered, the average number of sophomores who earned a passing score on their first attempt usually fell in a mid-60 percentage range.

“I have reason to doubt the validity of the apparent increase in the passing rate from the old exit exam to the new test,” Anderson said. “There’s something wrong here. It makes no sense"...

Anderson...calls the aversion to disclose the minimum level of performance required to pass the state’s exit exam “the dirty little secret of the testing business.” “I think a fair question to ask is: ‘What percentage of questions do you have to get correct to pass PACT or HSAP?’” Anderson said.

I can understand the state's reluctance to supply this answer, because (a) many testing companies don't disclose cutscores, and (b) it's quite likely that the test isn't scored with a simple number-right total. If item difficulty is also a factor, then an examinee with three hard items correct could end up with a higher score than someone with four easy items correct, and that's very hard to explain to laypeople. On the other hand, it's hard to defend oneself against the accusation that the test has been dumbed down if one isn't willing to release some information about how well someone has to perform to pass.

Posted by kswygert at April 18, 2005 09:35 AM
Sitemeter