Background:
Epidemiological studies have consistently found that women are typically 1.5 to 3 times more likely to experience major depression and about twice as likely to experience an anxiety disorder. One possible contributor is item or gender bias in the diagnostic tools and continuous assessments used to measure these constructs. The present study examines whether symptoms assessed by the Goldberg Anxiety and Depression Scales have a gender bias.
Methods:
A survey of 7485 people was carried out in the Canberra and surrounding Queanbeyan region (PATH Through Life Project). Confirmatory factor analysis was carried out to confirm the factor structure of the Goldberg scales and multiple group analysis was used investigate gender bias in the anxiety and depression items.
Results:
The results showed that several items were significantly gender biased. However, as the effects were small, their impact is likely to be minimal.
Conclusions:
It is important to investigate whether scales are gender biased when examining gender differences in anxiety and depression. This process helps to differentiate artifactual explanations from true differences.
