Decreased Scholastic Achievement in Overweight Middle School Students




The purpose of this study was to investigate whether overweight students achieved a lower relative degree of scholastic achievement compared to nonoverweight students. Subjects consisted of 6th and 7th grade students enrolled in a large public middle school in a suburb of Philadelphia, Pennsylvania. We compared grade point averages (GPAs), nationally standardized reading scores, school detentions, school suspensions, school attendance, tardiness to school, physical fitness test scores, and participation on school athletic teams among nonoverweight, at risk for overweight, and overweight students. Overweight students achieved lower grades (P < 0.001) and lower physical fitness scores (P < 0.0001) than their nonoverweight peers. Overweight students demonstrated a 0.4 letter grade lower GPA (on a 4.00 scale) and 11% lower national percentile reading scores than their nonoverweight peers. The overweight students also demonstrated significantly more detentions, worsened school attendance, more tardiness to school, and less participation on school athletic teams than their nonoverweight peers. Our study suggests that body mass is an important indicator of scholastic achievement, attendance, behavior, and physical fitness among middle school students, reiterating the need for healthy lifestyle intervention and prevention measures.


The percentage of overweight children in the United States has been on the rise for several decades (1). On the basis of estimates from the 2003–2004 National Health and Nutrition Examination Survey, 17% of the total population aged 12–19 years is overweight (2). Overweight children and adolescents are at risk for significant health problems, including insulin resistance, increased blood pressure, asthma, high cholesterol, obstructive sleep apnea syndrome, and psychosocial problems (3,4,5,6,7). Recent studies have provided a growing body of research for understanding the impact of pediatric obesity on health-related quality of life (8,9,10). In school domains, overweight students report being teased more, have lower health-related quality of life in social and academic realms, and have more school absenteeism than their nonoverweight peers (8,9,10,11,12,13). There are long-term consequences of pediatric obesity because overweight adolescents are more likely to complete fewer years of formal education, generate lower incomes, and have a greater likelihood of living in poverty as adults than their nonoverweight peers (14).

Although there is some evidence of decreased academic achievement among overweight students based on self-reported measures (15,16), this finding has not been demonstrated using objective school data. The purpose of this study was to determine whether there are significant differences in school achievement between overweight and nonoverweight middle school students as measured by objective school data. We hypothesized that a quantifiable achievement gap would be discovered between the overweight and nonoverweight students in this study.

Methods and Procedures


The study was approved by Temple University's Institutional Review Board and completed within the Review Board's guidelines. Full support of the study was granted by building and school district level administration. All data in this study were obtained through a secondary data analysis from existing school records of 6th and 7th grade students from a large public middle school in a suburb of Philadelphia, Pennsylvania. There were 271 6th graders and 301 7th graders enrolled in the school at the time of the study. Parents and guardians were notified in writing about the study. Parents were only required to notify us in writing if they did not consent to their child's records being included in the study. A total of six parents requested that their child's records be excluded from the study, leaving the records of 566 students to be analyzed in the study.

Dependent measures

Data used for this study were originally gathered by school personnel for purposes other than this study. Data were collected at the conclusion of the 2004–2005 academic school year. The dependent measures were placed into three categories including academic achievement, attendance and discipline, and physical fitness and athletic team participation.

Academic achievement measures. Grade point averages (GPAs) were collected from year-end student report cards. The GPAs reflect the cumulative average from the grades earned for each subject over the course of four 10-week marking periods. Weighted grades are not used in this school. Reading comprehension scores were obtained from the degree of reading power (DRP) test. This test was administered to all students during the first month of school by certified teachers for purposes other than this study. The DRP is a nationally standardized test that assesses reading comprehension of nonfiction texts. We recorded the national percentile score for all students.

Attendance and discipline measures. Year-end totals for the number of days absent, number of days tardy, number of administrative detentions assigned, and number of days suspended from school were recorded for all students.

Physical fitness and athletic team participation measures. Physical fitness data were collected from assessments that were conducted by certified physical education teachers during the school day in physical education classes in October. The assessments consisted of curl-ups, shuttle run, endurance 1-mile run/walk, pull-ups, and sit and reach. The purpose of the fitness tests was to determine students' eligibility for two nationally standardized physical fitness awards. The Presidential Physical Fitness Award is awarded to students who score at or above the 85th percentile (based on the 1985 School Population Fitness Survey) on all five tests. Students are awarded the National Physical Fitness Award for scoring above the 50th percentile (based on the 1985 School Population Fitness Survey) on all five tests (17).

Physical education teachers recorded the height and weight of all students using standardized measuring procedures concurrently with the physical fitness tests. We converted all heights and weights into BMI percentile scores using the 2000 Centers for Disease Control Weight by Age by Gender Tables (18). We placed each student into the nonoverweight (BMI% percentile <85), at risk for overweight (BMI% percentile 85–94), or overweight (BMI% percentile ≥95) category based on the Centers for Disease Control's parameters (18). We also recorded the participation of 7th grade students on school-based interscholastic athletic teams. Sixth grade students were not permitted to participate on school-based athletic teams.

Data analysis

Comparisons of means between the nonoverweight, at risk for overweight, and overweight groups were computed using a one-way ANOVA with Scheffe post hoc analysis. All tests were conducted controlling for demographic variables that included gender, socioeconomic status, and ethnicity. Socioeconomic status was determined by students' enrollment in the Federal Free and Reduced Lunch Program. All statistics are reported as mean ± s.d. For binary response variables, logistic regression was used (also adjusted for the demographic variables).


There were 406 nonoverweight, 85 at risk for overweight, and 58 overweight students. Sample characteristics are detailed in Table 1. Comparisons of the academic, achievement, and attendance and discipline variables for nonoverweight, at risk for overweight, and overweight students are listed in Table 2. For all indices of academic, achievement, and attendance and discipline, with the exception of suspensions and DRP national percentile scores, there were significant differences between overweight students and nonoverweight students when controlling for demographic variables. Specifically, the GPAs of nonoverweight students were ∼11% higher than those of the overweight students (P < 0.001). There was a general tendency for lower DRP national percentile scores in overweight students, compared to nonoverweight students, when controlled for demographic variables (although the difference between groups was not statistically significant). When detentions were categorized into groups of students who had 0–5 detentions and those who had 6 or more detentions, there was again a statistical difference between the nonoverweight and overweight students when controlling for demographics (P < 0.05). Overweight students were five times more likely to have 6 or more detentions than nonoverweight students. The rate of detentions in the nonoverweight students was one-half of that observed in overweight students when controlling for demographics. Nonoverweight students also had fewer absences (25%) and fewer days tardy to school (39%) relative to overweight students (P < 0.05). There were no significant differences between the at risk for overweight students and the nonoverweight students on the academic, achievement, and attendance and discipline measures. There was a trend for the at risk for overweight students to demonstrate better performance than overweight students on GPA (P = 0.08), but not on DRP scores or attendance and discipline measures.

Table 1.  Sample characteristics
inline image
Table 2.  Academic achievement, discipline, and attendance
inline image

The overall performance of the students on the measures of physical fitness is outlined in Table 3. As the table indicates, nonoverweight students performed better than their at risk for overweight and overweight peers when controlling for demographic variables (P < 0.0001). As expected, at risk for overweight students also performed at higher levels than overweight students on all fitness standards with the exception of the sit and reach test. The most notable differences in performance among all three groups were on weight-dependent tasks such as the pull-up, shuttle run, and 1-mile run.

Table 3.  Physical fitness test percentile scores
inline image

Figure 1 illustrates student participation on school athletic teams. As the figure highlights, 75% of all nonoverweight 7th grade students participated in at least one school-based athletic team compared to only 61% of at risk and 33% of the overweight students (P < 0.001, adjusted for demographics). This establishes an inverse relationship between school-based athletic team participation and BMI percentile.

Figure 1.

Percentage of school athletic team participation. This figure illustrates the differences in school athletic team participation rates among all three groups of students. Seventy-five percent of all nonoverweight 7th grade students participated in at least one school-based athletic team compared to only 61% of at risk and 33% of the overweight students (P < 0.001).


The major finding of this study was that nonoverweight students demonstrated better grades, a tendency toward higher reading scores, better attendance, and less school discipline relative to overweight youth. Existing literature has already found an inverse relationship between school achievement and BMI using self-reported measures of achievement and attendance (15,16). Our study expands the current literature because we reported on objective measures of student achievement. In addition, we reported on variables not commented on in the existing literature. The inclusion of data on discipline, reading comprehension, and participation in school athletics provides a more comprehensive profile of student achievement.

Overweight children score lower on measures of self-concept and engage in more high-risk behavior than nonoverweight children, which could interfere with scholastic functioning (19,20). It is possible that lower indices of self-concept and more high-risk behavior impact school achievement, attendance, behavior in school, and participation in school athletics. We speculate that decreased school attendance has a direct impact not only on the other dependent measures, but also on some factors that we did not measure such as peer relations, relations with teachers, and student satisfaction with school.

We were not surprised to find that nonoverweight students were more physically fit than overweight students, yet we were surprised to find that school-based athletic team participation was greater for nonoverweight students. This was an unexpected finding because this particular middle school has a “no-cut” policy (except for basketball) for sport participation and strongly encourages sport participation regardless of students' athletic ability. This trend is supported by a recent study that suggested overweight children entering adolescence have a low self-perception of their ability to participate in sports (21).

Interestingly, weight status significantly affected the performance of all three groups of students on the physical fitness measures, but only impacted the differences on the academic and attendance and discipline measures for the overweight and nonoverweight students. This suggests that the protective psychosocial factors such as self-efficacy and resiliency appear to be less compromised for academic tasks than physical tasks for students that are at risk for overweight. The at-risk students' elevated weight status might be a relatively recent development and thus, not have affected their psychosocial protective factors to the point of limiting their academic functioning. It would be interesting for future work to assess whether physical fitness declines at a faster pace than academic performance once a student reaches the overweight threshold.

Future research that builds on our work has the potential to bring more attention to the achievement gap noted in this study. Additional research in this area should consider some of the limitations of our study. Our study is limited by the inclusion of students from only one school, a narrow age range, and data collection from only one time interval. We recommend that future studies measure similar dependent variables on a larger and more diverse sample that includes students at the elementary and high school levels. A longitudinal study measuring these variables would help determine the long-term impact of pediatric overweight on school achievement.


Our findings add to the existing body of literature that suggest that further efforts are needed to address the issue of pediatric obesity. Schools have considerable potential to help students adopt healthy behaviors and can be effective vehicles of systemic change because almost 50 million students in the United States are enrolled in public schools (22,23). Local and State Boards of Education can evaluate the amount of physical activity, quality of health education, and the nutritional choices provided for students.

Specifically, it is suggested that middle school students receive 225 min of physical education per week that is based on rigorous national standards (24,25). Quality health education in middle school consists of adequate instructional time, addresses healthy eating, increases physical activity, and promotes less sedentary activities (25). Reducing foods that are high in fat or deep-fried and limiting access to vending machines are considerations for improving nutritional choices (26).

In summary, our study suggests that body mass is an important indicator of scholastic achievement, attendance, behavior, and physical fitness among middle school students, reiterating the need at this stage of development for healthy lifestyle intervention and prevention measures.


The authors declared no conflict of interest.