The instrument in question is the Children’s Response Styles Questionnaire, or CRSQ, a 25-item self-report scale adapted from the adult Response Styles Questionnaire. Developed originally in Canada by Abela and colleagues, the CRSQ asks young people what they typically do and think when they feel sad, with each item rated on a four-point scale from ‘Almost never’ to ‘Almost always’. Items such as ‘Think about how alone you feel’ tap into rumination, while items such as ‘Watch TV or play video games so you don’t think about how sad you are’ capture distraction and problem-solving. Although the questionnaire has been validated in samples from the United States, Turkey, Hong Kong and its native Canada, no one had systematically tested whether it works as intended in British adolescents — a gap that matters, because words and concepts can shift meaning across cultures even within the same language.

To close that gap, researchers led by Olivia Adams of the University of Oxford, together with Patrick Smith of King’s College London, Bao Sheng Loe of the University of Cambridge, and Felicity Waite and Eleanor Leigh of Oxford, pooled data from two school-based surveys conducted in South London. Of 763 young people aged 11 to 16 who took part, 615 provided complete responses to the CRSQ and formed the analytical sample. The group had a mean age of 13.32 years and was composed of 421 boys, 325 girls and 17 participants who did not report their gender. Notably, the sample was ethnically diverse: among those recording ethnicity in the first dataset, more than half identified as Black, Black British, Black Welsh, Caribbean or African — a proportion far exceeding the 4.2 percent recorded in the 2021 England and Wales census — making the study a valuable test of the measure in a population often underrepresented in psychometric research.

The analytical strategy was thorough. Because the questionnaire’s items are answered on an ordinal four-point scale, the team used weighted least squares estimation within a confirmatory factor analysis framework, testing whether the data supported the two-factor structure suggested by earlier studies: one factor for rumination and a combined factor for distraction and problem-solving, abbreviated DPS. Before pooling the two school datasets, the researchers confirmed through multigroup analysis that the questionnaire measured the same constructs in both. Bartlett’s test of sphericity and the Kaiser-Meyer-Olkin index, which came in at a strikingly high 0.91, confirmed the data were well suited to factor analysis in the first place.

The initial two-factor model fit imperfectly, and the refinement process proved revealing. Five items were dropped, all of them from the DPS subscale: ‘Spend a lot of time on schoolwork’, ‘Do something you enjoy’, ‘Do something fun with a friend’, ‘Read a book or magazine’ and ‘Ask a friend, parent or teacher to help you solve your problem’. The researchers suggest a common thread: unlike most other items, these make no explicit reference to feeling sad or to an event that triggered sadness. A conscientious student, for instance, might spend a lot of time on schoolwork for reasons having nothing to do with mood, diluting the item’s connection to the underlying construct. After removing the weak items and allowing two help-seeking items to share correlated error terms, the final model achieved strong fit, with a Comparative Fit Index of 0.985 and a Root Mean Square Error of Approximation of 0.062.

Reliability figures reinforced the asymmetry between the two halves of the scale. The rumination subscale showed excellent internal consistency, with Cronbach’s alpha exceeding 0.9, while the DPS subscale, though still above the conventional 0.7 threshold, was noticeably weaker — unsurprising given that five of its items had been removed. The researchers are candid about the implication: future use of the CRSQ with adolescents may be best restricted to the rumination subscale alone, with any analyses involving the DPS treated as exploratory until the subscale’s structure can be replicated. This echoes earlier concerns that the distraction construct, inherited from the adult questionnaire, may be poorly operationalised, bundling together effective and ineffective coping strategies.

A central question for any questionnaire intended for diverse groups is whether it functions identically across them — the property psychometricians call measurement invariance. Using multiple-group confirmatory factor analysis for categorical data, the team compared boys and girls, and separately compared younger and older participants split at the median age of roughly 13.3 years. For age, full measurement invariance was achieved across configural, threshold, metric and scalar levels, meaning the questionnaire measures rumination the same way whether a respondent is 11 or 15. For gender, the scalar model initially faltered, but freeing thresholds for two items restored adequate fit, yielding partial invariance — sufficient, the authors note, for comparing mean scores between boys and girls without serious concern that the construct itself is being measured differently.

The invariance findings proved substantively interesting in their own right. Girls scored higher than boys on the rumination subscale, replicating a well-documented gender difference that emerges in adolescence and that previous research has linked to the widening depression gap between the sexes. Older participants also ruminated more than younger ones, a statistically significant difference consistent with the idea that rumination intensifies across the teenage years. Meanwhile, the two factors correlated modestly and positively with each other, a pattern the authors attribute to shared variance in mood-related responding rather than any meaningful overlap between dwelling and distraction.

Convergent validity — whether the questionnaire’s scores track related constructs as theory predicts — was examined against four established symptom measures: the Children’s Depression Inventory and the Multidimensional Anxiety Scale for Children in one dataset, and the Mood and Feelings Questionnaire and the Screen for Child Anxiety Related Disorders in the other. Rumination scores correlated moderately and positively with both depression and anxiety symptoms, and crucially, these correlations survived statistical controls. The link between rumination and anxiety remained significant after accounting for depression, and the link between rumination and depression remained significant after accounting for anxiety, with partial correlations of 0.36 and 0.45 respectively. This pattern supports the emerging view of rumination as a transdiagnostic process — a single cognitive habit that feeds into multiple forms of emotional distress rather than depression alone.

The study is not without limitations, which the authors enumerate with unusual thoroughness. The sample came from a single South London borough, so replication across more rural areas and the other UK nations is needed; ethnicity data were missing for one dataset; socioeconomic information was absent; and only binary gender was recorded. The cross-sectional design precludes test-retest and predictive validity checks, and no independent measure of rumination was available to strengthen convergent validity. The combination of two datasets, while statistically justified, means independent replication would be welcome. Most importantly, the sample was a community one, and given rumination’s proposed role as a target for both prevention and treatment, testing the measure in clinical populations — ideally alongside diagnostic interviews rather than self-report alone — is the natural next step.

Even so, the implications are considerable. Depression affects between seven and twenty-four percent of adolescents worldwide, raises the risk of adult depression nearly threefold, and undermines physical health and academic attainment. If repetitive negative thinking is indeed a modifiable mechanism in this cascade, then researchers and clinicians need instruments they can trust to detect it. This study provides the first UK-specific evidence that the CRSQ’s rumination subscale is both reliable and valid in British teenagers, invariant across age and largely invariant across gender, and meaningfully connected to the symptoms it should predict. The caution attached to the distraction and problem-solving subscale is itself a service to the field, steering future studies away from a component that may mislead. As the authors conclude, ensuring that measures are psychometrically robust is a quiet but essential foundation for the larger project of understanding — and eventually interrupting — the cognitive pathways that carry young people into depression.

Subject of Research: Psychometric validation of the Children's Response Styles Questionnaire for measuring adolescent rumination in UK youth

Article Title: Psychometric Validation of the Children's Response Styles Questionnaire in UK Adolescents

Article References: Adams, O., Smith, P., Loe, B. S., Waite, F., & Leigh, E. (2026). Psychometric Validation of the Children's Response Styles Questionnaire in UK Adolescents. Child Psychiatry & Human Development. https://doi.org/10.1007/s10578-026-02084-6

Image Credits: AI Generated

DOI: 10.1007/s10578-026-02084-6

Keywords: rumination, adolescent depression, psychometrics, Children's Response Styles Questionnaire, measurement invariance, UK adolescents, anxiety symptoms, Response Styles Theory, confirmatory factor analysis, mental health, repetitive negative thinking, questionnaire validation