The Strengths and Difficulties Questionnaire (SDQ): the Factor Structure and Scale Validation in U.S. Adolescents

Abstract

The Strengths and Difficulties Questionnaire (SDQ) is one of the most commonly used instruments for screening psychopathology in children and adolescents. This study evaluated the hypothesized five-factor structure of the SDQ and examined its convergent validity against comprehensive clinical diagnostic assessments. Data were derived from the National Comorbidity Survey - Adolescent Supplement (NCS-A), a nationally representative sample of U.S. adolescents aged 13 to 18 years. Parents/parent surrogates (n=6,483) was asked to complete a self-administered questionnaire including the SDQ and DSM-IV comprehensive diagnostic information on the participating adolescents. Confirmatory factor analysis (CFA) was conducted to assess the factor structure of the SDQ. The five-factor solution of the SDQ (including emotional, conduct, hyperactivity-inattention, peer relationship, and prosocial) provided a satisfactory fit to the data, and was invariant across sex, age, race/ethnicity and income subgroups. SDQ scores predicted a significantly increased probability of meeting criteria for a DSM-IV disorder, with better prediction for behavior disorders than for mood disorders. Decreasing the SDQ cutoffs to the 80th percentile significantly increased the sensitivity from 39% to 63% for the SDQ Total Difficulties Score, with an expected decrease in specificity from 93% to 87%. This work confirms the five-factor structure of the SDQ in an ethnically and sociodemogrpahically diverse community sample of adolescents. Our findings strengthen empirical evidence for the use of the parent-reported SDQ as a screening tool for DSM-IV behavioral and emotional disorders in adolescents identified in the general population.

New evidence of factor structure and measurement invariance of the SDQ across five European nations

“…The study of the internal structure, by means of CFAs, supported the five-factor structure in all the countries and in the total sample, as it is the case in previous studies [40,27,11,24,28,29,25].…”

“…Previous studies, conducted using confirmatory factor analysis, showed the five-factor solution as the most appropriate [27,24,11,28,29,25], while others found the three-factor model [30][31][32]11]. The threefactor solution is composed by Internalizing symptoms, resulting of the Emotional and Peer problems subscales, the Externalizing symptoms, composed by Conduct problems and Hyperactivity subscales, and the Prosocial subscale.…”

“…As yet, there has been no in-depth examination addressing the question of whether the dimensional structure underlying the SDQ scores is invariant across countries. To the best of our knowledge, few studies have addressed this question [39,38,27]. For instance, Essau et al [38], with a sample of 2418 adolescents, found that the factorial structure of the SDQ differed across five European countries (Cyprus, England, Sweden, Germany, and Italy).…”

Scale quality: alpha is an inadequate estimate and factor-analytic evidence is needed first of all

“…In case of a multidimensional construct, we assume that the scale is split into subscales. For example, the Strengths and Difficulties Questionnaire (SDQ) consists of five subscales (He, Burstein, Schmitz, & Merikangas, 2013). Subsequently, omega is reported per subscale.…”

Life skills program for improving adolescent mental health in the af-termath Mount Merapi eruption, Yogyakarta-Indonesia

“…(Youthinmind Ltd, 2015) As the participants in this study were adolescents (junior and senior high school students), the SDQ version used in this study was the self-report version. This instrument has been widely used worldwide (Gómez-Beneyto, 2013;He, 2013;Liu, 2013), as well as in Indonesia in the previous study (Ang, 2007).…”