Psychometric properties of the Swedish version of the Childhood Trauma Questionnaire-Short Form (CTQ-SF).

Childhood maltreatment is delicate to assess both in clinical work and in research. There is a need for assessment tools that can be easily administered in an ethical and non-intrusive way that meets requirements of conceptual validity for various types of maltreatment and is sensitive to levels of severity. This study explores the psychometric properties of the Swedish translation of one such tool-the Childhood Trauma Questionnaire-Short Form (CTQ-SF; Bernstein and Fink, 1998). The CTQ-SF was administered to seven samples (total n=659)-five clinical samples and two non-clinical student samples. The factor structure supports the construct validity of the global maltreatment scale, four of the five maltreatment subscales (emotional abuse, physical abuse, sexual abuse and emotional neglect) and the minimization/denial (MD) scale, but not the physical neglect (PN) subscale. All items are highly correlated with their respective subscale. The discriminant validity is satisfactory. Highly significant correlation with social desirability gives further support for the MD-scale and to the recommendation of how to apply it. Internal consistency of PN is acceptable and for all other scales satisfactory. Swedish norm groups tend to score lower than similar American norm groups on abuse scales but higher on the neglect scales. Percentiles for seven gender-specific norm groups are presented. The weaknesses of the PN-scale are discussed and new constructs are proposed. The Swedish version of the CTQ-SF has the same construct validity and internal consistency as the original, including less homogeneity of the PN scale.