Elsevier

Annals of Epidemiology

Volume 15, Issue 4, April 2005, Pages 272-278
Annals of Epidemiology

Factors Associated with Errors in Self-reports of Stature, Weight, and Body Mass Index in Minnesota Adolescents

https://doi.org/10.1016/j.annepidem.2004.08.010Get rights and content

Purpose

Obtaining stature, weight, and body mass index (BMI) from self-reports rather than by direct measurements is highly attractive economically and logistically, but there are few data available for adolescents that allow evaluation of potential sources of reporting bias. Because self-reports are based on self perceptions, personal characteristics of youth may be related to errors in reporting.

Methods

Differences between self-reported and measured stature, weight, and BMI were investigated for a sample of 3797 Minnesota youth, 12 to 18 years of age. Gender, age, race/ethnicity, socioeconomic status (SES), and measured body size were examined as potential factors associated with errors in self-reports.

Results

Self-reported stature, weight, and BMI were generally highly correlated with corresponding measured dimensions, although adolescents of both genders systematically overestimated their statures, underestimated their weights, and underestimated BMI. Significant associations of errors in self-reports with age, race/ethnicity, and socioeconomic status suggested that differences in self perceptions or value ascribed to body size varies according to these personal characteristics. Systematic errors in self-reported stature, weight, and BMI were negatively associated with the corresponding measured dimension when adjusted for age, race/ethnicity and SES, such that prevalences of overweight based on BMI from self-reported measures were systematically underestimated relative to measured values.

Conclusions

Self-reports of stature, weight, and BMI are on the average, valid representations of their measured counterparts; nevertheless, errors in the self-reports are systematically related to characteristics of youth. Consequently, findings from these studies should be interpreted carefully.

Introduction

Adolescent obesity is a major public health concern (1) and there is great interest in collecting data on overweight and obesity in a wide range of settings (2). Obtaining stature, weight, and body mass index (BMI) from self-reports rather than by direct measurements is highly attractive economically and logistically. Of course, the appropriateness of using self-reports depends on the validity, reliability, and specific applications of these measures.

There is considerable literature for adults concerning self-reported stature, weight, and BMI. Overall, corresponding self-reported and measured dimensions are highly correlated 3, 4, 5, 6. Even so, rather consistent biases occur in self-reported dimensions. Self-reported stature generally overestimates measured stature, especially in males and the elderly, while self-reported weight, on the average, underestimates measured weight, and consequently BMI 7, 8. At the extremes of the distributions of measured size, the biases in self-reported body size are toward the mean, and especially so for large weights and BMI 9, 10, 11, 12.

Although self-reported measurements are commonly used in studies on adolescents and in major surveys such as the US national Youth Risk Behavior Survey (13), there are relatively few studies of the validity of self-reported stature and weight in youth. Most validation studies are based on rather small and specialized populations 14, 15, 16, 17, 18, 19, 20 and have inconsistent findings and limited generalizability. In general, findings indicate that adolescents overestimate stature by approximately 0.5 to 1.0 cm, and underestimate weight by about 1 to 2 kg, relative to measured dimensions. Results from one large study of youth in grades 9 to 12, however, showed an average overestimation of stature of 2.7 cm and underestimation of weight by 3.5 kg (21). Results from another large data set (NHANES III), however, depart substantially from this general pattern and indicate underestimation of stature by 0.3 to 3.0 cm within each age group in both males and females (22). For the NHANES youth, the largest mean underestimations in reported stature were at 12 and 13 years of age, and non-Hispanic Black youth tended to underestimate stature to a greater degree than non-Hispanic White or Mexican American children.

Few of the available studies in adolescents have been sufficiently large or diverse to allow investigation of individual characteristics that may be associated with errors in self-reporting stature, weight, and BMI. Understanding the potential influences of individual characteristics such as gender, age, race/ethnicity, and socioeconomic status are important for designing studies and for interpreting data where self-reported stature and weight are used. Moreover, variation in errors related to personal characteristics may give insight into the awareness, perceptions, and social implications of body size in adolescents. Reported herein are findings concerning overall validity and selected factors associated with errors in self-reported stature, weight, and BMI from a large sample of youth attending schools in Minnesota (23). The differences between self-reported and measured dimensions are considered errors in self-reports and patterns and associations of these errors with gender, age, race/ethnicity, socioeconomic status, and the measured body size of participants are investigated.

Section snippets

Methods

The sample includes 3797 youth (1936 males, 1861 females), 12 to 18 years of age, who had complete data for measured and self-reported stature and weight collected in the course of Project EAT (Eating Among Teens), a comprehensive study of adolescent nutrition and obesity 23, 24. Body measurements were not obtained on 10.6% of youth targeted for the study, largely because of absence from school on data-collection days. Self-reported stature and weight were missing in 5% and 4%, respectively, of

Results

The number of cases and relative proportions of the sample (total n = 3797) according to child characteristics are presented in Table 1. The uneven distribution across the whole-year age groups results from primarily targeting youth attending grades seven and ten. The distribution according to race/ethnicity reflects a slight majority white population at the participating schools. The distribution of SES demonstrates substantial socioeconomic variation in the sample.

Table 2 presents the Pearson

Discussion

The results from these analyses confirm that the overall validity of youth reports of stature, weight, and BMI is high, with Pearson correlations from 0.80 to 0.96 and intraclass correlations from 0.77 to 0.96. These levels are generally consistent with results from other studies 18, 20, 21.

Nevertheless, when one looks beyond the average associations, one finds that the differences between self-reported and measured stature, weight, and BMI are systematically related to selected characteristics

References (35)

  • A.L. Stewart

    The reliability and validity of self-reported weight and height

    J Chronic Disease

    (1982)
  • R.P. Troiano et al.

    Overweight prevalence and trends for children and adolescents. The National Health and Nutrition Examination Surveys, 1963 to 1991

    Arch Pediatr Adolesc Med

    (1995)
  • J.H. Himes et al.

    Reported versus measured adult statures

    Am J Phys Anthropol

    (1982)
  • K. Nakamura et al.

    Reliability of self-reported body height and weight of adult Japanese women

    J Biosoc Sci

    (1999)
  • H. Payette et al.

    Validity of self-reported height and weight estimates in cognitively-intact and impaired elderly individuals

    J Nutr Health Aging

    (2000)
  • J.C. Alvarez-Torices et al.

    Self-reported height and weight and prevalence of obesity. Study in a Spanish population

    Int J Obes Relat Metab Disord

    (1993)
  • C. Bolton-Smith et al.

    Accuracy of the estimated prevalence of obesity from self reported height and weight in an adult Scottish population

    J Epidemiol Community Health

    (2000)
  • Cited by (182)

    • Association between lifestyle-related, psychosocial factors and obesity among female adolescents in Taiwan

      2023, Journal of Pediatric Nursing
      Citation Excerpt :

      However, self-reported height and weight have been found to be highly correlated with objectively measured height and weight in community samples (Pursey et al., 2014). Moreover, self-reported height and weight have been observed to be well correlated with measured height and weight among adolescents (Berge et al., 2015; Himes et al., 2005). Furthermore, the prevalence of overweight and obesity observed in our study is in accordance with those reported in previous studies (Ni et al., 2019; Park, 2011a).

    View all citing articles on Scopus

    This work was supported by grant MCJ-270834 (D. Neumark-Sztainer, principal investigator) from the Maternal and Child Health Bureau (Title V, Social Security Act), Health Resources and Service Administration, US Department of Health and Human Services.

    View full text