Reliability and validity of the youth asset survey (YAS)

• Published in: Journal of adolescent health 2002-09, Vol.31 (3), p.247-255
• Main Authors: Oman, Roy F, Vesely, Sara K, Mcleroy, Kenneth R, Harris-Wyatt, Vicki, Aspy, Cheryl B, Rodine, Sharon, Marshall, Ladonna
• Format: Article
• Language: English
• Subjects: Adolescent; Adolescent Behavior; Adolescent Development; Adolescent risk behaviors; Adolescents; Biological and medical sciences; Demography; Ethnic differences; Factor Analysis; Factor Analysis, Statistical; Female; Gender differences; Humans; Internal-External Control; Interview, Psychological; Interviews as Topic; Male; Measures (Instruments); Medical sciences; Psychological tests; Psychology. Psychoanalysis. Psychiatry; Psychometric Analysis; Psychometric properties; Psychometrics; Psychometrics. Diagnostic aid systems; Psychopathology. Psychiatry; Reliability; Risk; Risk behaviour; Risk-Taking; Surveys; Surveys and Questionnaires; Techniques and methods; USA; Validity; Young people; Youth; Youth Asset Survey; Youth assets; Youth development; Youth development programmes
• ISSN: 1054-139X; 1879-1972
• DOI: 10.1016/S1054-139X(02)00363-4

Purpose: To describe the construction of a survey that was developed to measure youth assets. The paper details the ten youth developmental assets measures within the survey, including the psychometric properties of those measures derived from factor analysis and reliability testing. Methods: Three studies were conducted, including a study of 1350 randomly selected youth (mean age = 15.4 years; 52% female; 47% white, 22% black, 19% Hispanic, 10% Native American), using in-person interviews, to develop an asset instrument tool. Factor analysis was performed on a correlation matrix, using principal axis factoring, and varimax rotation. The criteria were set to extract eight factors. Results: The eight factors extracted represented six developmental assets as originally defined, including Family Communication, Peer Role Models, Future Aspirations, Responsible Choices, Community Involvement, and Non-Parental Role Models. One asset (Constructive Use of Time) was split into two specific assets: groups/sports and religious time. Two assets did not form factors (Good Health Practices [exercise/nutrition] and Cultural Respect) and were defined as one-item assets. All factor loading scores were .40 or higher and all Cronbach alphas were .60 or higher. Conclusions: Factor analyses suggest that the constructs are reliable measures of youth developmental assets. The psychometrically sound asset measures presented here will provide scientists with valid and reliable instruments to assess and compare the prevalence of youth assets across populations and to investigate potential relationships between youth assets and other outcomes, such as youth risk behaviors.