Excargodog

All majors don't use the Hogan. Yet those that don't do not appear to have pilots substantially inferior to those that do. That would seem to scuttle the business necessity argument.

Why are you so anxious to defend a test that has little if any power to distinguish between people who will be successful in the job and those who won't?

The issue comes down to something in group statistics called Bayes Theorem. If you were using personality testing on an unscreened group - just guys off the street - personality testing like MMPI Or the Hogan would have reasonable statistical power to pick out the 1% who are schizophrenic and the 3 or 4% who are chronically depressed. The predictive value of a positive assessment or negative assessment might be pretty decent. But that isn't what you are doing. When you change the population being screened to a group that - for instance, has safely accumulated 5000 flying hours including 1500-2000 of 121 time, has passed a half dozen physicals (even FAA physicals), has a clean police record, passed the KCM check, and has a few recommendations from people with similar qualifications who know him/her, you've ALREADY eliminated the vast majority of the problem children, so even if you apply the exact same criteria, the value of the test goes way down.

Your percentage of false positives doesn't change, but there are barely any true positives to be found. Basically if you use the results to cull half the applicants, the bottom half that you are eliminating do not differ in any meaningful way from the 'top' half you are keeping.

I mean, don't take my word for it. Look up Bayes Theorem or ask some statistician. By the time you apply a screening test to a population that has already been heavily screened by other criteria you might as well just flip a coin and save the money.


https://betterexplained.com/articles...bayes-theorem/