N Lien1, C Friestad, K I Klepp. 1. Institute for Nutrition Research, University of Oslo, PO Box 1046, Blindern, 0316 Oslo, Norway. nana.lien@basalmed.uio.no
Abstract
STUDY OBJECTIVE: Children's proxy reports on indicators of their parents' socioeconomic status (SES) have either been used uncritically or dismissed as invalid. This paper examines the validity of young adolescents' reports of parental SES by comparing adolescent reports with parents' own reports of SES. DESIGN, SETTING AND PARTICIPANTS: In 1990, 924 13 year olds, along with 648 of their fathers and 735 of their mothers, participated in the baseline survey of The Norwegian Longitudinal Health Behaviour Study. Data on parental occupation were collected from both adolescents and parents at baseline, while data on parental education were collected at follow up from the adolescents at age 15 and from their parents when the adolescents were 19 years old. MAIN RESULTS: Three different ways of grouping the SES categories based on occupational data were investigated, and the strength of agreement was good for all three groups, with kappa statistics ranging from 0.65 to 0.86. There were no significant improvements of agreement when comparing adolescent data from age 15 to adolescent data from age 13. The strength of agreements between the adolescents' and parents' reports of parental education were fair; kappa statistics were 0.30 and 0.38 for fathers' and mothers' education, respectively. The proportions of unclassified answers or no responses from the adolescents were similar for questions on occupation and education, and ranged from 11% to 16%. CONCLUSIONS: The agreement between adolescents' and parents' reports of SES based on occupation was judged to be good, but adding a few specific questions may cue the adolescent to provide more detailed information, thereby reducing the numbers of unclassified answers or non-responders.
STUDY OBJECTIVE:Children's proxy reports on indicators of their parents' socioeconomic status (SES) have either been used uncritically or dismissed as invalid. This paper examines the validity of young adolescents' reports of parental SES by comparing adolescent reports with parents' own reports of SES. DESIGN, SETTING AND PARTICIPANTS: In 1990, 924 13 year olds, along with 648 of their fathers and 735 of their mothers, participated in the baseline survey of The Norwegian Longitudinal Health Behaviour Study. Data on parental occupation were collected from both adolescents and parents at baseline, while data on parental education were collected at follow up from the adolescents at age 15 and from their parents when the adolescents were 19 years old. MAIN RESULTS: Three different ways of grouping the SES categories based on occupational data were investigated, and the strength of agreement was good for all three groups, with kappa statistics ranging from 0.65 to 0.86. There were no significant improvements of agreement when comparing adolescent data from age 15 to adolescent data from age 13. The strength of agreements between the adolescents' and parents' reports of parental education were fair; kappa statistics were 0.30 and 0.38 for fathers' and mothers' education, respectively. The proportions of unclassified answers or no responses from the adolescents were similar for questions on occupation and education, and ranged from 11% to 16%. CONCLUSIONS: The agreement between adolescents' and parents' reports of SES based on occupation was judged to be good, but adding a few specific questions may cue the adolescent to provide more detailed information, thereby reducing the numbers of unclassified answers or non-responders.
Authors: Charlotte Meilstrup; Lau Caspar Thygesen; Line Nielsen; Vibeke Koushede; Donna Cross; Bjørn Evald Holstein Journal: Int J Public Health Date: 2016-02-03 Impact factor: 3.380