Does weight status affect academic performance? Evidence from Australian children

ABSTRACT

This article investigates the impact of children’s weight status on their academic performance using the Longitudinal Survey of Australian Children. Considering the endogeneity of the weight status variable (obesity or overweight), to obtain consistent estimates of its impact, we use the body mass index of biological parents as instrumental variables. The two-stage least square estimation shows that obesity or overweight has a significant negative impact on academic performance. Furthermore, the effect of child weight status is different across grade levels. Overweight has a larger negative impact on academic performance for senior year students, especially on numeracy.

KEYWORDS:

• Obesity
• overweight
• academic performance
• instrument variable
• Australia

JEL CLASSIFICATION:

• I1
• I2
• C1

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Non-cognitive outcomes are valued by households and teachers, creating a bias that affects an objective reflection of children’s skills, so these are not considered in this study.

² Overweight and obesity are measured at the population level using the BMI which is calculated by dividing weight in kilograms by height in metres squared. Overweight and obesity for Australian children and adolescents is determined by comparing calculated BMI against the relevant age and gender of the child/adolescent.

³ Relative disadvantage is a technical term used by the ABS to summarise information about the economic and social conditions of people and households within an area, including the Index of Relative Socio-economic Advantage and Disadvantage (IRSAD).

⁴ We did not attempt an FE approach because only 6% of children’s weight status changed during the study period, which makes FE results from this small subsample of little interest.

⁵ http://www.health.gov.au/internet/main/publishing.nsf/content/health-pubhlth-strateg-hlthwt-obesity.htm.

⁶ From the data we have no obese children from the East Asia background according to the Australian classification.

⁷ Empirical studies have long noted the existence of an ‘Asian effect’ on students’ academic performance (see Chen and Stevenson 1995; Fejgin 1995; Huang and Waxman 1995; Kao 1995; Schneider and Lee 1990). Sun (1998) builds an investment model to study racial differences in educational performance. It is indicated that families from East Asian origins invest more aggressively in financial, human and within-family social capital than families from other racial groups. These differences in investment explain a large portion of the Asian effect. Byun and Park (2012) also assess the relevance of shadow education to the high academic performance of East Asian American students by examining how East Asian American students differed from other racial/ethnic students in the prevalence, purpose, and effects of using commercial test preparation service and private one-to-one tutoring.

⁸ Averett and Stifel (2010) and Scholder et al. (2012) also find that the coefficients in the IV models are much larger than in the OLS and FE models when mother’s past BMI are used as an instrument.

⁹ Falsification tests are never definitive but can be helpful in demonstrating the sensitivity and specificity of the model (Cawley and Meyerhoefer 2012).

¹⁰ When using English-speaking background as the benchmark, the variable ‘eastasia’ still shows significantly positive impact on five subjects.

Additional information

Funding

This work was supported by the Victoria University [CRGS 15/15].