Tests are fine if they test performance
By Raymond Dreyfack -- Plant Engineering, 3/1/1999
When Grade I Electrician Art Roland opted for early retirement, Maintenance Supervisor Harry Crane pulled out the roster. Two names stood out as likely candidates for promotion from Grade II to Grade I: Pete Wilcox and George Johnson. From an experience standpoint both men appeared qualified. The question was: Which one is most qualified?
He gave each candidate two tests: a general intelligence test, and a test that related specifically to the work. Wilcox scored 84% on the performance test to Johnson's 83%. But his score on the intelligence test was 13% higher than Johnson's. That decides it, Crane thought.
Within minutes of posting his choice, Johnson appeared at Crane's desk. He wanted to know why, since their performance scores were almost the same, and since he had better seniority than Wilcox, he didn't get the job.
"Wilcox did much better on the IQ test," Crane replied.
"That doesn't mean beans," Johnson said. "That test is obsolete and has no bearing on the job's performance."
Crane shrugged. "You're entitled to your opinion."
"Darn right I am," Johnson replied sullenly.
Question: Does Johnson have an actionable case against Crane's decision?
Belcher's verdict: When Plant Engineer Frank Belcher learned about the intelligence test on which Crane had based his decision, he told the supervisor it was as outmoded as button down shoes and instructed him to discontinue using it.
"Johnson's black, isn't he?"
"Yes, but that has nothing to do -- "
"I'm sure it doesn't, but it's important to realize that these ancient general tests are all the more unfair in that they give Caucasian workers an edge. Tests can be useful, but only if they test performance as it applies to the job."
