> then it is almost surely that the study was a correlation between IQ and some questionnaire result

No, because that once again assumes IQ tests are a relevant way to measure intelligence.