The objective of this study was to assess item characteristics indicative of the severity of risk for commercial sexual exploitation among a high-risk population of child welfare system involved youth to inform the construction of a screening tool. Existing studies have discerned factors that differentiate Commercial Sexual Exploitation of Children (CSEC) victims from sexual abuse victims, yet no research has been conducted to discriminate which items in a high risk population of youth are most predictive of CSEC. Using the National Survey of Child and Adolescent Well-Being (NSCAW) cohorts I and II, we examined responses from 1063 males and 1355 females ages 11 and older, over three interview periods. A 2-parameter logistic Item Response Theory (2 PL IRT) model was employed in order to examine item performance as potential indicators for the severity of risk for CSEC. Differential Item Functioning (DIF) analysis was conducted in order to examine potential differences in item responses based on gender. Modeling strategies to assess item difficulty and discrimination were outlined and Item Characteristic Curves for the final retained items were presented. Evidence for uniform DIF were present within items that asked about runaway, any drug use, suicidality, and experiencing severe violence. Results from this study can inform the construction of a screening instrument to assess the severity of risk for experiencing CSEC. © 2018 Elsevier Ltd

DIF Item response theory sex trafficking Differential item functioning Risk Commercial sexual exploitation of children Human trafficking maltreatment IRT Child welfare system involvement

human factor analysis risk assessment clinical assessment tool violence controlled study interview male female prediction Child Welfare high risk population sexual exploitation victim child sexual abuse commercial phenomena Article runaway behavior suicidal behavior gender cohort analysis disease severity statistical model Child