That's my point -- the HR monkey was a social scientist, and the fact that they were administering a test and getting a number out gave them the illusion that they were actually doing something scientific.
An actual scientist would have asked questions -- Do the test results correlate with performance? Is the test biased? What is the margin of error on the test? etc. -- and would very likely have come to the conclusion that the test should not be trusted as a reliable indicator.
