Back to Annual Meeting

Back to Annual Meeting

APHA Scientific Session and Event Listing

Suzanne L. Slocum, PhD, Philip Batterham, MPH, Heather Tevendale, PhD, and Marguerita Lightfoot, PhD. Center for Community Health, Semel Institute for Neuroscience and Human Behavior, University of California Los Angeles, 10920 Wilshire Blvd., Suite 350, Los Angles, CA 90024, 310-794-6270, SSlocum@mednet.ucla.edu

Background: Homeless youth often engage in a broad range of problem behaviors, including delinquency, mental health problems, substance use, and sexual risk behaviors. Because these behaviors often occur concurrently, they can be conceptualized and examined as a single constellation of risk, termed multiple problem behaviors. Although various problem behaviors are often examined as a singe summary variable, little is known about the psychometric properties and the validity of the inferences made from multiple problem behavior (MPB) scores. The authors of the current study investigated the construct validity and psychometric properties of a measure of MPB for runaway youth. Method: 407 homeless youth aged 12-20 were recruited from service agencies and followed longitudinally for twelve months. An MPB score was calculated by examining unprotected sex, problems with alcohol and marijuana use, hard drug use, conduct disorders, school problems and delinquent behaviors. The MPB score was assessed for item convergent validity, item discrimination, internal consistency reliability, test-retest reliability, and predictive validity using baseline and follow-up data. Results: Results indicate an essential unidimensional model for MPB. Although sexual risk had a low factor loading on MPB ( 0.37), eigenvalue tests and parallel analysis indicated a one factor model. All internal consistency coefficients were >0.60, and the MPB score was found to be a significant predictor of several HIV outcomes in follow-up data. Conclusions: The MPB measure exhibited strong psychometric properties. Results suggest that the MPB measure is a reliable and valid measure appropriate for use in studies of homeless and at-risk youth.

Learning Objectives: At the conclusion of the session, the participant (learner) in this session will be able to

• Identify a list of behaviors that represent a single constellation of HIV risk for homeless adolescents.
• Evaluate procedures that are used to test the validity and reliability of a self-report instrument.
• Develop a score that represents multiple problem behaviors.

Presenting author's disclosure statement:

Not Answered