PURPOSE: The study's purpose was to improve the psychometric properties of the Youth Asset Survey (YAS). DESIGN: Longitudinal cohort study with youth and parents recruited via door-to-door canvassing to participate in a 5-wave, 4-year study that assessed prospective associations among youth assets and youth health-related behaviors. Additional test items were added in the last 2 waves of a study to improve the YAS. SETTING: Households in randomly selected census tracts located in the Oklahoma City metropolitan area. PARTICIPANTS: Youth and their parents (N = 1111 parent/child dyads). Youth demographic characteristics at wave 4 were mean age = 17.3 (standard deviation = 1.62) years; 53% female, 41% white, 28% Hispanic, 24% African American, and 6% other. MEASURES: Sixty-eight items assessing 17 youth asset constructs and 8 single items assessing youth risk behaviors. ANALYSIS: Confirmatory factor analysis and generalized linear models were conducted to assess construct reliability and predictive validity, respectively. RESULTS: Cronbach α for the revised asset constructs ranged from .72 to .82, predictive validity was strong, and all revised asset constructs were assessed via 4 items. CONCLUSIONS: Researchers and practitioners have an improved 68-item YAS-Revised, freely available for their use, which measures 17 youth assets with good reliability, validity, and functionality.
PURPOSE: The study's purpose was to improve the psychometric properties of the Youth Asset Survey (YAS). DESIGN: Longitudinal cohort study with youth and parents recruited via door-to-door canvassing to participate in a 5-wave, 4-year study that assessed prospective associations among youth assets and youth health-related behaviors. Additional test items were added in the last 2 waves of a study to improve the YAS. SETTING: Households in randomly selected census tracts located in the Oklahoma City metropolitan area. PARTICIPANTS: Youth and their parents (N = 1111 parent/child dyads). Youth demographic characteristics at wave 4 were mean age = 17.3 (standard deviation = 1.62) years; 53% female, 41% white, 28% Hispanic, 24% African American, and 6% other. MEASURES: Sixty-eight items assessing 17 youth asset constructs and 8 single items assessing youth risk behaviors. ANALYSIS: Confirmatory factor analysis and generalized linear models were conducted to assess construct reliability and predictive validity, respectively. RESULTS: Cronbach α for the revised asset constructs ranged from .72 to .82, predictive validity was strong, and all revised asset constructs were assessed via 4 items. CONCLUSIONS: Researchers and practitioners have an improved 68-item YAS-Revised, freely available for their use, which measures 17 youth assets with good reliability, validity, and functionality.
Entities:
Keywords:
adolescents; age specific; measurement issues; psychometrics; research methods; specific populations; youth asset instrument; youth asset measurement; youth development
Authors: Vivienne M Hazzard; Alison L Miller; Katherine W Bauer; Bhramar Mukherjee; Kendrin R Sonneville Journal: J Adolesc Health Date: 2019-11-10 Impact factor: 5.012