Determining agreement between self-reported and measured heights and weights in 8th- and 11th grade Texas public school students

Few studies have examined the agreement between self-reported and measured heights and weights in adolescents at different grade levels in an ethnically diverse sample. The aim of this research was to determine if self-reported height and weight yielded reproducible measures against measured equivalents and to estimate to what degrees self-reported measures deviated from measured values in Texas school children. Data from the School Physical Activity and Nutrition survey, a cross-sectional probability-based survey conducted from 2000-2002, were used. Eighth and 11th grade students in Texas (n = 8497, 55.3% male; 53.4% 8th grade; 33.9% Hispanic, 10.5% African American) were asked to report their height and weight; afterwards, these students were weighed and measured by study staff. Agreement between self-reported and measured height, weight, and BMI was assessed using Lin's concordance correlation coefficient. Agreement ranged from 0.79-0.91 in 8th-and 0.92-0.96 in 11th grade students. No substantial differences occurred across gender and race/ethnicity; across BMI categories, agreement measures were widely dispersed (8th:0.18-0.55; 11th:0.31-0.78). Height measures were less reliable than weight measures. Height was generally over-reported and weight was under-reported for both grades. Height was more likely to be over-reported in 11th grade students compared with 8th grade students. Errors in self-reporting were prominent at extreme ends of measured height, weight, and BMI, especially in 8th grade students. Self-reported measures in adolescents should be cautiously employed in research studies where the outcome is body mass index, particularly at younger ages, due to reporting bias.