Investigating Perceptions of the Eating Disorder Examination-Questionnaire Among Undergraduate Students: A...

Area of Science:

• Psychology
• Psychometrics
• Eating Disorder Research

Background:

• Quantitative evaluations of the Eating Disorder Examination Questionnaire (EDE-Q) have consistently shown that its original four-factor structure does not replicate across diverse samples and genders.
• Qualitative methods offer nuanced insights into how individuals interpret and respond to self-report survey items, potentially explaining variations in Likert scale responses and gendered experiences.

Purpose of the Study:

• To qualitatively explore how lay participants perceive and interpret the items within the Eating Disorder Examination Questionnaire (EDE-Q).
• To identify potential misunderstandings or ambiguities in EDE-Q items that may affect response validity across different genders.

Main Methods:

• 97 undergraduate students (48 men, 48 women, 1 transgender man) responded to the 22 attitudinal EDE-Q items using a Likert-type scale.
• Participants provided open-ended reflections on their thought processes, interpretations, and contextual factors influencing their response selection.
• Content analysis was used to examine open-ended responses, focusing on item and scale comprehension, including confusion, misinterpretation, validity concerns, and gendered experiences.

Main Results:

• While most responses were clear, participants frequently conflated items related to shape and weight, expressing confusion.
• Higher Likert scale ratings did not consistently correlate with greater eating psychopathology.
• Men reported eating psychopathology related to weight gain and muscularity, even when providing low Likert scale responses.

Conclusions:

• Qualitative findings from this study corroborate concerns about the structural validity of the EDE-Q raised by quantitative research.
• The results highlight essential concepts within the EDE-Q that may be misunderstood, suggesting areas for clarification and improvement in the questionnaire.