Overview
The CRSQ is a 25-item self-report measure which asks what young people do and think in response to feeling sad. Respondents answer on a 4-point Likert scale from 'Almost Never' to 'Almost always'. Items are summed to produce an overall CRSQ scale score, a rumination subscale score, and a combined distraction and problem-solving subscale score. The CRSQ was adapted from a questionnaire for adults, the Response Styles Questionnaire (RSQ, Nolen-Hoeksema & Morrow, 1991). The original version is a 71-item questionnaire with an added scale for Dangerous Activities. The CRSQ has been translated and validated in Turkish and Chinese samples (Ozguluk et al., 2012; Lo et al., 2017).Scoring & Interpretation
Responses for each item are assigned a number from 0 to 3, where responses of 'Almost never' are scored 0 and responses of 'Almost always' are scored 3. Due to the two-factor structure of the measure, responses are summed, such that responses to the odd-numbered questions contribute to a rumination subscale score, and responses to the even-numbered questions contribute to the combined distraction and problem-solving score. Higher scores indicate higher use of the response style. Some papers calculate a ratio score, by dividing the rumination subscale score by the distraction and problem-solving subscale score. There are no reverse-scored items.Psychometric Properties
Internal Consistency
Moderate levels of internal consistency have been reported, with an early paper citing internal consistency for the rumination subscale = .82, and for the distraction and problem-solving scale = .79 (Abela, Aydin & Auerbach, 2007). Other research has indicated the distraction and problem-solving subscale has internal consistency = .67 (Ozguluk et al., 2012).
Test-Retest Reliability
Over 4 weeks, the test-retest reliability was .72 for rumination and .71 for the distraction and problem-solving subscale, indicating acceptable test-retest reliability.
Factor Structure
Initially, early use of the CRSQ treated the three subscales (Rumination, Distraction and Problem-Solving) separately, although acknowledged that as rumination increases, scores for other scales decrease (Abela et al., 2002). However, since then, factor analyses have provided a two-factor solution, where the distraction and problem-solving subscales are combined, and a ratio calculated (division of the rumination score by the combined distraction and problem-solving scores, Abela et al., 2007). Confirmatory factor analysis supported the two-factor model, (CFI=.98, RMSEA=0.049, Abela et al., 2007).
Validity
Convergent validity has been demonstrated: Positive correlations were found between rumination subscale scores and depression symptoms scores (r = .27 to .44). Negative correlations were found between distraction and problem-solving scores and depression symptom scores (r = -.28 to -.31) (Ozguluk et al., 2012).