Assessing the Construct Validity and Reliability of School Health Records Using the ‘Health Dialogue Questionnaire’ in the Eleventh Grade

Background and aims The aim for this study was to assess the construct validity and reliability of the Health Dialogue Questionnaire (HDQ©) for eleventh grade in school through comparison of the HDQ© with Paediatric Quality Of Life Inventory (PedsQL™), Local monitoring of youth policy questionnaire (LUPP®), Health behaviour in Swedish school-aged children (HBSC©), Equal health (EH©) and The Swedish Survey Youth on Alcohol Consumption (SSYAC©). Methods Cross-sectional samples of eleventh graders from the academic year 2009/2010 was used from the HDQ© (n = 2752), the HBSC© (n = 2090), the PedsQL™ (n = 666), the "LUPP®" questionnaires (n = 2400), EH© (n = 258), and SSYAC© (n = 1748) in the academic year 2009/2010. A comparison between HDQ© and the different proxies was done. Exploratory and confirmatory factor analyses were performed as well as a Multitrait-Multimethod Matrix (MTMM), in order to evaluate the construct validity and reliability of HDQ©. Results An average disagreement between HDQ© and proxies with 10 percentages was found. Exploratory factor analysis of HDQ© on the 2009/2010 sample suggested a four factor solution (girls factor solution 65% of total variance explained, and in the boys' solution 59% of total variance explained). A second sample 2010/2011 of eleventh graders were used for the confirmatory solution. Almost perfectly similar four factor solutions with were found (girls 58% of total variance explained and boys 56% of the total variance explained). Using MTMM the reliability was generally high and HDQ© and showed agreeable validity. Discussion and conclusions The HDQ© questionnaire is a reliable and valid instrument for measuring eleventh graders self-reported-health in school.