The Emotional Eating Scale adapted for children and adolescents: Factorial invariance across adolescent males and females

• The metric and stricter invariance models needed to have constraints removed on the worried and bored items.
• Caution should be used when including the worried and bored item in the EES total score and subscale scores.
• These item may have different meaning for adolescent girls and boys.

ObjectiveThe purpose of the present study was to assess factorial invariance of the Emotional Eating Scale (EES) across a sample of adolescent males and females.MethodTwo-hundred-twenty-six middle school students (mean age = 13.19 years; SD = 1.00) completed the EES adapted for children and adolescents and a brief demographic survey. Approximately 40% of the sample were males. An item-level multi-group confirmatory factor analysis was performed specifying a bi-factor model that mapped onto the EES' four scores.ResultsWe found that the EES was partially invariant across sex in the current study's sample. The instrument appeared to measure the same constructs in both groups, except for the two items: worried and bored.ConclusionsOur findings suggest that overall the EES items measure their intended constructs in both male and female adolescents, but caution should be used when interpreting scores that include the worried and bored items. These two item may have be measuring their constructs differently in adolescent females and males.