There was no evidence of substantive DIF for any item by either gender or age. One item, however, had moderate DIF by gender (item 6), and two items (12 and 16) had moderate DIF by age, although each in the opposite direction thus canceling their impact at the scale (total score) level. It should be noted that if we had applied the Bonferroni correction for multiple comparisons, no item would have been identified with DIF.
Criterion validity. Sensitivity, specificity, and ROC curve. The scores obtained on the symptom questionnaire ranged from 0 to 24. A good balance between sensitivity and specificity was found for a cutoff of 6, with values of 75.0% and 70.2%, respectively. A lower cutoff point would increase the sensitivity, but many workers without CVS would be