Have States Reduced Obesity by Legislating More Physical Activity in Elementary School? (2024)

Journal List
Milbank Q
v.101(1); 2023 Mar
PMC10037682

As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice

Milbank Q. 2023 Mar; 101(1): 204–248.

Published online 2023 Mar 13. doi:10.1111/1468-0009.12604

PMCID: PMC10037682

PMID: 36913506

PAUL T. von HIPPEL¹ and DAVID E. FRISVOLD²

Author information Copyright and License information PMC Disclaimer

Abstract

Policy Points

Between 1998 and 2016, 24 states and the District of Columbia passed laws meant to increase the time that children spent in physical education (PE) or other school‐based physical activity (PA).
Schools largely disregarded changes to PE/PA laws, which did not increase the time that children spent in PE or recess and did not reduce body mass index, overweight, or obesity.
Closer oversight of schools would be needed to improve compliance with state PE and PA laws. Yet, even with better compliance, we estimate that PE and PA policies would be inadequate to reverse the obesity epidemic. Policies should also address consumption, both inside and outside of school.

Context

To control childhood obesity, leading medical organizations have recommended increasing the time that children spend in physical education (PE) and other school‐based physical activity (PA). Yet, it is unknown how many states have passed laws that codify these recommendations, and it is unknown what effect changing state laws has had on obesity or the time that children actually spent in PE and PA.

Methods

We joined state laws to national samples of 13,920 children from two different cohorts of elementary students. One cohort attended kindergarten in 1998; the other attended kindergarten in 2010; both cohorts were followed from kindergarten through fifth grade. We estimated the effects of changes to state laws in a regression with state and year fixed effects.

Findings

Twenty‐four states and the District of Columbia increased the time that children were recommended or required to spend in PE or PA. These changes in state policies did not increase actual time spent in PE or recess, did not affect average body mass index (BMI) or BMI Z score, and did not affect the prevalence of overweight or obesity.

Conclusion

Increasing the PE or PA time required or recommended by state laws has not slowed the obesity epidemic. Many schools have failed to comply with state laws. A back‐of‐the‐envelope calculation suggests that, even with better compliance, the legislated changes in PE laws might not have changed energy balance enough to reduce obesity prevalence.

Keywords: physical education and training, pediatric obesity, schools, exercise, policy, cohort studies

Leading medical organizations, including the american Academy of Pediatrics (AAP),1 the American Heart Association (AHA),2 and the Institute of Medicine (IOM),3 have recommended increasing the time that children spend in physical education (PE) and other school‐based physical activity (PA). The AAP has recommended that schools require daily PE in all grades,1 the AHA has recommended that students participate in at least 30 minutes/day (150 minutes/week) of school‐based PA,2 and the IOM has recommended that elementary students spend 30 minutes/day (150 minutes/week) in PE alone, with additional PA in recess, classroom activities, and extracurricular activities.3

Although PA can have a number of benefits for physical and mental health, both the AAP and the AHA highlighted obesity prevention as a primary goal of their PA policy recommendations.1 Their rationale was straightforward. Increases in PA can increase energy expenditure (calories out), which should cause weight loss, or reduce weight gain, if there is no change in energy consumption (calories in). PA may also affect appetite, satiety, and resting metabolic rate, although the evidence for those effects is less clear.4, 5

Despite this rationale, a 2013 article in the New England Journal of Medicine classified the claim that PE classes reduce obesity as a “myth.”6 A 2019 Cochrane Review of 14 randomized controlled trials (RCTs) enrolling 16,410 children aged 6–12 years concluded that interventions that increased PA without modifying diet reduced average body mass index (BMI) by only 0.1 kg/m² (95% confidence interval [CI]: 0.05 to 0.14),7 or approximately 0.18 kg (0.40 lb) for a 9‐year‐old of average height (1.33 m). In 2016, the AAP, which had previously described the evidence from RCTs as “somewhat disappointing to date,”1 announced that it had retired its policy statement endorsing PA to prevent child obesity.8 The IOM and the AHA, however, have not changed their recommendations.

Although numerous RCTs have evaluated the obesity effects of specific PA interventions, it is just as important to understand what has been accomplished by state PApolicies. Unlike an RCT, in which investigators can closely monitor schools’ fidelity in implementing the details of a designed intervention, state policies typically set guidelines and leave the details of implementation to schools and districts. With little state supervision, there is a risk that schools will either ignore state policy or implement it in a way that is unlikely to affect obesity—in addition to the risk that even a high‐fidelity implementation might still have little effect.

The most common state policies are laws that recommend or require a minimum time for PA or PE. Evaluations of state policies for PA and PE time have produced mixed results. A national study of students attending high school in 1999–2003 found that in states with PE time requirements, students spent an average of 31 minutes/week longer in PE than students in other states, but the extra PE time was not associated with lower BMI or lower prevalence of overweight and obesity.9 A national study of students attending elementary school in 1998–2005 found that compliance with state PE time requirements was poor; in states with PE time requirements, only 17% of schools met the requirement in kindergarten, and 45% met the requirement in fifth grade.10 Nevertheless, elementary students in states with higher PE requirements did spend more time in PE than students in other states; moreover, the extra PE time was associated with lower BMIs and lower prevalence of overweight and obesity—but only among boys, and only in fifth grade.10 A later study of elementary students in 2010–2016, however, found no association between the extra PE time associated with state policies and BMI, overweight, or obesity.11

To date, evaluations of state PA/PE time policies have focused on short periods when policies did not change, comparing students in states that had higher time requirements with students in states that had lower time requirements or no time requirements at all. Although the studies controlled for numerous observed covariates (including race and ethnicity, parental income and education, and participation in school meal programs), the estimates remained vulnerable to bias from omitted variables that might confound the relationship between BMI and state PE time policies. In addition, because past evaluations focused on short time periods, they could not directly address the question of what happens when a state increases the time that it recommends or requires for school‐based PA or PE.

In this article, we estimate the effects of increases in state PE and PA requirements. These are our specific research questions:

Between 1998–1999 and 2015–2016, how many states increased the time that they legislated for PE and PA in elementary school?
Did elementary schools comply with legislated increases? That is, did the time that children actually spent in PE and PA increase with changes in state laws?
Were increases in legislated PE and PA time associated with reductions in BMI, obesity, or overweight, net of other variables?

Methods

We correlated changes in elementary students’ outcomes—PE time, recess time, BMI, obesity, and overweight—with changes in state laws for PE and PA time between 1998 and 2016.

We describe our data, research design, and statistical model below. All our code and data are available on the website of the Open Science Foundation (https://osf.io), except for data from the Early Childhood Longitudinal Studies, Kindergarten (ECLS‐K) cohorts, which are available to anyone with a restricted data license from the National Center for Education Statistics.

Data on Laws for PE and PA Time

We gathered data on the state PE and PA time laws that were in effect between 1998 and 2016. Our sources were the Lexis Library Legal Database,12 the State Policy Database published by the National Association of State Boards of Education,13 and the District of Columbia Official Code.14 We searched for changes to state laws pertaining to elementary school “physical education” or “physical activity” between 1998 and 2016, as well as older laws, passed as early as 1940, which were in effect in 1998. All state laws that were in effect during the period of the study are described in the Appendix Appendix33 Table.

The language used in state laws was inconsistent. Some laws addressed PE, and others addressed PA; we treated those as distinct. Some laws set “requirements,” whereas others made “recommendations;” we treated those as distinct as well. Some laws asked schools to “offer” PE, whereas others asked that students “enroll” in PE; we treated those terms as equivalent. Most laws specified the number of minutes per week, but three laws (in Wisconsin, Illinois, and Connecticut) specified the days of PE per week; we transformed days into minutes by assuming 30 minutes of PE per day.

Data on Children and Schools

We merged state policy data with data from the Early Childhood Longitudinal Studies of the Kindergarten Classes of 1998–1999 (ECLS‐K:1999) and 2010–2011 (ECLS‐K:2011).15 The ECLS‐K are longitudinal cohort studies overseen by the US Department of Education, National Center for Education Statistics. Each of the two ECLS‐K studies followed a different cohort of children longitudinally. The first study began with children who attended kindergarten in 1998–1999 (ECLS‐K:1999); the second began with children who attended kindergarten in 2010–2011 (ECLS‐K:2011).

Both ECLS‐K studies started with nationally representative probability samples of the US kindergarten population. Both studies followed children for at least 6 years. Both studies used a multistage clustered sample design, sampling children within schools and sampling schools within primary sampling units, each of which was a large county or a group of adjacent and demographically similar small counties.

Together, the two ECLS‐K cohort studies contained data on 39,590 students in 7,670 schools. After restricting the data to children with complete data on the variables needed for our analysis, we had an analytic sample of 13,920 students in 3,000 schools. Details of the sample restrictions are given in Table1.

Table 1

Summary of Sample Restrictions and Effective Sample Size

Analytic Sample	Students	Schools
Combine ECLS‐K:1999 and ECLS‐K:2011	39,590	7,670
Restrict to students observed in the fifth grade round	23,790	4,790
Restrict to students with nonmissing BMI in the fifth grade round	21,980	4,490
Restrict to students with nonmissing BMI in fall kindergarten	19,380	4,060
Restrict to students with nonmissing demographic characteristics	19,110	4,030
Restrict to students with nonmissing family characteristics	16,990	3,760
Restrict to students with nonmissing PE variables	15,470	3,280
Restrict to students with nonmissing recess variables	13,920	3,000

Open in a separate window

Abbreviations: BMI, body mass index; ECLS‐K:1999, Early Childhood Longitudinal Studies, Kindergarten Class of 1998–1999; ECLS‐K:2011, Early Childhood Longitudinal Studies, Kindergarten Class of 2010–2011; NCES, National Center for Education Statistics; PE, physical education.

Data are derived from ECLS‐K:1999 and ECLS‐K:2011. Sample sizes are rounded to the nearest ten to comply with NCES confidentiality requirements.

School PE and PA Time. Both ECLS‐K cohort studies measured children's weekly participation in PE and recess. In each grade, teachers were asked how many days per week children participated in PE and how many minutes children participated on those days. We analyzed the responses from fifth grade teachers. Instead of reporting the exact amount of time spent in PE or recess, teachers chose intervals from a list of time intervals. The intervals for minutes/day differed across the two cohorts: the intervals were 0, 1–15, 16–30, 31–60, and more than 60 minutes/day in the 1999 cohort, but 0, 1–29, 30–59, 60–89, 90–119, 120–149, 150–179, and more than 180 minutes/day in the 2011 cohort. Because the intervals were not the same, it was challenging to compare PE time across the two cohorts, but we estimated changes in the continuous underlying time variable using regression models appropriate for interval‐censored data (decribed later).

Recess was measured in days/week, and fifth grade teachers could give any response from 0 to 5. Fifth grade teachers were asked one more question about the time that children spent in recess, but the question was not the same for the two cohorts—it asked about minutes in the earlier cohort and sessions in the later cohort—so we did not use it in our analyses.

BMI, BMI Z Score, Overweight, and Obesity. In both ECLS‐K cohorts, trained survey personnel measured children's heights and weights on at least seven occasions between the fall of kindergarten and the spring of fifth grade. Personnel asked students to remove shoes, hats, and jackets, then measured height using a Shorr board stadiometer and weight using a digital scale. Children were measured twice on each occasion, and the two measurements were averaged after any major discrepancies were resolved.

From height in meters (m) and weight in kilograms (kg), we calculated BMI (kg/m2). Then, following guidelines defined by the Centers for Disease Control and Prevention, we calculated a BMI Z score (zBMI), as well as dummy variables indicating whether each child exceeded age‐ and gender‐specific thresholds for overweight or obesity.16 We also defined a dummy variable for extreme obesity, defined as a BMI exceeding 1.2 times the age‐ and gender‐specific threshold for obesity.17

Our primary outcomes were BMI, zBMI, overweight, and obesity at the end of fifth grade. Fifth grade measurements reflected the cumulative effect of PE and PA policies across all 6 years of elementary school (kindergarten through fifth grade). 92% of children finished fifth grade on schedule (in spring 2004 for the earlier cohort, or spring 2016 for the later cohort). The other 8% finished in different years, typically because they had repeated a grade, less often because they had skipped one. We included all outcomes measured in spring 2004 or 2006, with a control variable for the grade that children were in if was not fifth grade. Before fifth grade data were released, we used third grade BMI as an outcome and obtained similar results.

Covariates. Covariates available in both ECLS‐K cohorts included race and ethnicity (five categories), gender, household income (thousands of dollars per year), household size (number of residents in the household), and parental education (years of education for the most educated parent). Although additional covariates are imaginable, we were practically limited to covariates that were available in the ECLS‐K and measured in the same way for both cohorts. Fortunately, our research design had features that reduced the danger of unobserved variables confounding the estimates.

Research Design

Our study was observational, but three features of our research design substantially reduced the danger of confounding by unobserved variables, bringing us closer to being able to interpret the results causally.

First, we used a difference‐in‐differences design, one of the oldest quasi‐experimental designs used to estimate causal effects in epidemiology.18 Our difference‐in‐differences design compared changes in “treatment” states—which, in the time between cohorts, passed laws intended to increase PE/PA time—with changes in “control” states, which did not pass such laws. We asked whether schools in treatment states increased PE/PA time more (or decreased it less) than schools in control states. We also asked whether fifth grade BMI, zBMI, overweight, and obesity decreased more (or increased less) in treatment states than in control states.

The advantage of a difference‐in‐differences design is that it holds many variables constant, even if they cannot be observed. Our design, in particular, held constant all differences between states that did not change over time (for example, children being heavier in Southern states), as well as time trends that did not vary between states (for example, if the later cohort were heavier than the earlier cohort, though it was not, according to Table2). A difference‐in‐differences design can estimate the effect of a policy change under the assumption of “parallel trends.”19 Here, the parallel trend assumption means that if no states had changed policies, the outcome (e.g., BMI) would have changed at similar rates in states that did in fact change policies and in states that did not.

Table 2

Descriptive Statistics of Student‐Level and School‐Level Characteristics

	1999 cohort		2011 cohort
Variables Measured in Spring of Fifth Grade	Mean	(SD)	Mean	(SD)
BMI	20.48	(4.68)	20.33	(4.74)
zBMI	0.61	(1.10)	0.58	(1.13)
Overweight	0.39	(0.49)	0.38	(0.49)
Obesity	0.21	(0.40)	0.21	(0.41)
PE days/week	2.11	(1.22)	1.83	(1.46)
Recess days/week	4.68	(0.90)	4.82	(0.71)
PE minutes/week required	30.38	(47.36)	41.72	(54.24)
PE minutes/week recommended	2.58	(10.47)	10.42	(36.25)
PA minutes/week required	1.16	(4.98)	32.87	(55.90)
PA minutes/week recommended	0.00	(0.00)	2.60	(15.90)
Age (months)	134.58	(4.42)	133.10	(4.40)
Female	0.49	(0.50)	0.49	(0.50)
Grade	4.91	(0.31)	4.94	(0.25)
Black	0.09	(0.29)	0.08	(0.26)
Hispanic	0.17	(0.38)	0.24	(0.43)
Asian	0.07	(0.25)	0.07	(0.25)
Other race/ethnicity	0.06	(0.23)	0.06	(0.23)
Household income (thousands of dollars)	68.61	(56.77)	89.84	(69.61)
Household size	4.59	(1.38)	4.68	(1.37)
Parents’ education (years, for the more educated parent)	14.59	(2.70)	14.75	(2.84)
Variables Measured in Fall of Kindergarten
BMI (fall of kindergarten)	16.33	(2.20)	16.50	(2.39)
Age (months, fall of kindergarten)	68.47	(4.35)	67.56	(4.40)
Children	7,750		6,170

Open in a separate window

Abbreviations: BMI, body mass index; PA, physical activity; PE, physical education; SD, standard deviation; zBMI, BMI Z score.

Data sources for state PE and PA laws are the same as in Figure1. Data sources for other variables were the Early Childhood Longitudinal Studies, Kindergarten Classes of 1998–1999 (ECLS‐K:1999) and Early Childhood Longitudinal Studies, Kindergarten Classes of 2010–11 (ECLS‐K:2011).

A further advantage of our research design is that we controlled for BMI at the start of kindergarten. Although it is not possible to control for every imaginable influence on BMI, BMI at the start of kindergarten represented the cumulative effect of all out‐of‐school influences that affected BMI before elementary school, some of which may also have affected BMI during elementary school and during summer vacations. Controlling for a lagged or "pretreatment" measure of an outcome variable, such as BMI, can reduce bias in causal estimates by 80% or more.20

A final feature of our research design was that we controlled for a variety of child and family covariates, some of which are associated with overweight and obesity, and some of which changed at different rates in different states. For example, the Hispanic population grew faster in some states than in others, and Hispanic children h higher overweight prevalence‐‐so we controlled for Hispanic ethnicity in our model. In our main analyses, we controlled for all the covariates listed in the Data on Children and Schools section above. In supplementary analyses, reported in the Appendix Appendix11 Table, we controlled for three additional covariates, and the results did not change.

Statistical Model

We fit our difference‐in‐differences design with a regression model that included state fixed effects, cohort fixed effects, and various control variables. For example, to model the effects of state laws on BMI, we fit the following regression:

$\begin{matrix} B M I_{5, s c i} & = & α_{s} + β_{c} + γ_{1} P E T i m e R e q_{s c} + γ_{2} P A T i m e R e q_{s c} \\ + γ_{3} P E T i m e R e c_{s c} + γ_{4} B M I_{K, s c i} + γ_{5} A g e_{5 s c i} + γ_{6} R a c e_{s c i} \\ + γ_{7} I n c o m e_{s c i} + γ_{8} P a r E d_{s c i} + e_{s c i} \end{matrix}$

Here, $B M I_{5, s c i}$ was the BMI of child i from cohort c in state s in the spring of fifth grade (or more precisely in the child's fifth year after kindergarten, when 92% of children were in fifth grade). The variables $P E T i m e R e q_{s c i}$ , $P A T i m e R e q_{s c i}$ , and $P E T i m e R e c_{s c i}$ were the times, in minutes per week, that state s required or recommended that children in cohort c spend in PE or PA. If a state had no law for PE or PA time when a cohort was in elementary school, then we coded the time recommended and required as 0. If a state changed laws while a cohort of children was in elementary school, we averaged that state's laws across that cohort's elementary years.

The parameters of interest were γ₁, γ₂, and γ₃, which represent the effects of PE and PA time laws on BMI. All the other terms controlled for potential confounders. State fixed effects $α_{s}$ controlled for differences among states that did not change with time. Cohort fixed effects $β_{c}$ controlled for differences between cohorts that did not vary among states. The most important control variable was the child's starting $B M I_{K, s c i}$ . The model also controlled for the child's age and grade in school at the time of the first and last BMI measurement, as well as race and ethnicity (five categories), household income (thousands of dollars per year), household size (number of residents in the household), and parental education (years of education for the most educated parent). The model also controlled for gender; supplemental analyses fit models separately by gender, with similar results; see Appendix Table Table2.2. The final term $e_{s c i}$ was a random residual, clustered at the school level to estimate standard errors robust to heteroscedasticity and autocorrelation.21

By changing the dependent variable, we used the same model to estimate the effects of changes in state laws on the BMI Z score (zBMI) or on dummies for overweight and obesity. We also estimated the effect of changes in state laws on the days/week spent in PE and recess, as well as the minutes/day spent in PE (minutes/day was not available for recess in the later cohort). Because minutes/day was interval‐censored and the intervals were not consistent across cohorts, we used interval regression with state fixed effects, as implemented by the intreg command for Stata.

Results

Changes to State Laws

Between 1998 and 2016, 24 states and the District of Columbia changed laws meant to increase the time that elementary students spent in PE or PA. Figure1 shows each state's changes with the fall that they became effective. Eleven states increased required PE time, and six states increased recommended PE time. Fourteen states increased required PA time, and just one state, Connecticut, increased recommended PA time. Increasing recommended PA time is the weakest possible policy change, but for Connecticut, it was only transitional; the state increased recommended PA time effective in fall 2004 and then increased required PA time effective in fall 2012. Because Connecticut's PA recommendation law was in effect for only two school years when ECLS‐K children were in elementary school (2010‐2011 and 2011–2012), we could not reliably identify the effect of the PA recommendation law and omitted it from our model.

Open in a separate window

Figure 1

Between 1998 and 2016, 24 States and the District of Columbia Passed Laws Increasing Requirements or Recommendations for PE or PA Time

Abbreviations: ECLS‐K, Early Childhood Longitudinal Studies, Kindergarten. Gray bands represent the elementary school years of the two cohorts of children in this study. Laws come from the Lexis Library Legal Database, the State Policy Database published by the National Association of State Boards of Education (NASBE),13 and the District of Columbia Official Code. This figure summarizes laws for fifth grade, but laws for other elementary grades were practically identical.

Figure2 graph trends in average recommended and required PE and PA time across all states that changed laws. In those states, average required PE time increased from 0 to 45 minutes/week, average recommended PE time increased from 3 to 27 minutes/week, and average required PA time increased from 0 to 67 minutes/week. These are average increases; in some states, the increases were more substantial. Sixteen states passed laws that, in some sense, met the 150 minutes/week target set by the IOM or the AHA: five states increased required PE time from 0 to 150 minutes/week; four states increased recommended PE time from 0 to 150 minutes/week, and seven states increased required PA time, including but not limited to PE, from 0 to 150 minutes/week.

Open in a separate window

Figure 2

Trends in PE and PA Time Recommended or Required in States that Changed Laws

Abbreviations: PA, physical activity; PE, physical education. In states that changed laws between 1998 and 2016, required PA increased time from 0 to 67 minutes/week, required PE time increased from 0 to 45 minutes/week, and recommended PE increased time from 0 to 27 minutes/week, on average. States that did not change laws required 27 minutes/week of PE on average and had no other requirements or recommendations. Data sources are the same as in Figure1. Averages give equal weight to every state.

In the 26 states that did not change laws, required PE time averaged 27 minutes/week, with a range of 0 to 150 minutes/week. Those states had no laws for PA time and no laws that recommended PE time without requiring it.

Obesity prevention was often an explicit goal of state PE and PA laws. Legislators and reporters discussed obesity in connection with the laws, and sometimes, the text of the law mentioned obesity explicitly. Texas's 2001 law, for example, stated that PE and PA were part of a coordinated school health program “designed to prevent obesity, cardiovascular disease, and type II diabetes in elementary school students.”22

Effects on Time Spent in PE and Recess

Contrary to the laws’ intent, increases in required or recommended PA/PE time did not increase the time that children actually spent in PE or recess. Table3 provides estimates from our difference‐in‐differences analysis. Among 12 coefficients estimating the impact of laws on different measures of time spent in PE or recess, five coefficients were nonsignificant (p > 0.05) and six coefficients were significant but negative, implying that increases in the PA/PE time required or recommended by law actually decreased the time that children spent in PE or recess. For example, increases in the PE time required or recommended by law predicted small but significant decreases in the days per week, and had no significant impact on minutes per day, that children actually spent in PE. Likewise, increases in PA time recommended by law predicted decreases in the PE days per week and PE minutes per day while having no significant impact on recess days per week.

Table 3

Difference‐in‐Differences Estimates of State Laws’ Effects on Fifth Graders’ Time in PE and Recess

	Fifth Grade Outcomes
	PE	PE	Recess
	(Days/Week)	(Minutes/Day)	(Days/Week)
PE minutes/week required	−.0047**	−.0391	.0047**
	(−.0083, −.0011)	(−.0864, .0081)	(.0015, .0080)
PE minutes/week recommended	−.0045***	−.0117	−.0008
	(−.007 −.0018)	(−.0533, .0298)	(−.0026, .0010)
PA minutes/week required	−.0073***	−.0854***	.0002
	(−.0101, −.0045)	(−.1245, −.0464)	(−.0011, .0015)
Age (spring fifth grade)	.0142	−.2511	.0057
	(−.0328, .0611)	(−.9652, .4631)	(−.0254, .0368)
Female	−.0056	−.0497	−.0081
	(−.0469, .0356)	(−.7315, .6321)	(−.0326, .0165)
Grade	.0739	.5443	−.1238***
	(−.0372, .1850)	(−1.1565, 2.2450)	(−.1870, −.0606)
Black	−.1634**	−.5913	−.2051***
	(−.2863, −.0405)	(−2.5210, 1.3384)	(−.3187, −.0916)
Hispanic	.0276	−.3416	−.0688
	(−.0760, .1312)	(−2.0597, 1.3765)	(−.1391, .0014)
Asian	−.1181	−.3155	−.0317
	(−.2383, .0022)	(−2.5778, 1.9468)	(−.0881, .0248)
Other race/ethnicity	−.1215*	−.6511	−.0245
	(−.2328, −.0103)	(−2.4647, 1.1624)	(−.0835, .0346)
Household income	.0002	.0146**	.0002
	(−.0003, .0007)	(.0056, .0236)	(−.0001, .0006)
Residents in household	.0150	−.1260	.0010
	(−.0016, .0317)	(−.4205, .1685)	(−.0094, .0113)
Parents’ education	−.0003	.1648	.0011
	(−.0113, .0106)	(−.0140, .3436)	(−.0059, .0080)
BMI in fall of kindergarten	.0016	−.0637	.0085**
	(−.0077, .0110)	(−.2132, .0857)	(.0031, .0139)
Age in fall of kindergarten	−.0119	.1697	−.0039
	(−.0593, .0355)	(−.5516, .8909)	(−.0353, .0275)
2011 cohort	−.0319	3.3453**	.0848*
	(−.1778, .1139)	(1.1422, 5.5484)	(.0149, .1547)
Children	13,920	13,920	13,920

Open in a separate window

Abbreviations: BMI, body mass index; PA, physical activity; PE, physical education.

***p < 0.001, **p < 0.01, *p < 0.05. School‐clustered 95% confidence intervals in parentheses. State policy variables are in bold. All models include state and cohort fixed effects. Data sources are the same as in Table2.

The coefficients look trivial at first glance, but they represent the effect of just a 1‐minute increase in time recommended or required per week. Many states increased required or recommended time by more than 100 minutes/week, and we can estimate their effects by multiplying the coefficients by 100. For example, a 100 minutes/week increase in required PE time reduced actual PE by 0.47 days/week (95% CI: 0.83 decrease to 0.11 decrease, p < 0.001), on average—a small but not a trivial effect.

Effects on BMI, Overweight, and Obesity

Changes in state laws had no effect on average BMI, average zBMI, or the proportion of children with overweight or obesity in the spring of fifth grade. Table4 shows estimates from our difference‐in‐differences analysis. Not only were the effects of state law changes not significant different from 0 (all p values > 0.05), the point estimates were near 0 with narrow CIs. For example, a 100 minutes/week increase in required PE was estimated to decrease the prevalence of overweight by just 0.3 percentage points (pp; 95% CI: 0.7 pp decrease to 0.1 pp increase, p > 0.05), increase the prevalence of obesity by just 0.1 pp (95% CI: 0.3 pp decrease to 0.4 pp increase, p > 0.05), reduce average zBMI by just 0.02 (95% CI: 0.10 decrease to 0.06 increase, p > 0.05), and reduce average BMI by just 0.06 kg/m² (95% CI: 0.35 decrease to 0.23 increase, p > 0.05)—or 0.1 kg (0.2 lb) for a 9‐year‐old of median height (1.33m). Even if these effects were statistically significant, which they were not, effects of this size would have no significance for children's health.

Table 4

Difference‐in‐Differences Estimates of State Laws’ Effects on Fifth Grade BMI, BMI Z Score, Overweight, and Obesity

	BMI	BMI Z Score	Overweight	Obesity	Extreme Obesity
PE minutes/week required	−.0006	−.0002	−.0003	.0001	.0001
	(−.0035, .0023)	(−.0010, .0006)	(−.0007, .0001)	(−.0003, .0004)	(−.0001, .0003)
PE minutes/week	−.0020	−.0004	−.0002	−.0003	−.0000
recommended	(−.0056, .0017)	(−.0012, .0005)	(−.0007, .0003)	(−.0006, .0001)	(−.0002, .0002)
PA minutes/week required	−.0003	.0001	−.0001	−.0000	−.0000
	(−.0026, .0020)	(−.0005, .0007)	(−.0004, .0002)	(−.0003, .0002)	(−.0001, .0001)
Age	.0472	−.0065	−.0033	.0009	.0016
	(−.0067, .1011)	(−.0216, .0086)	(−.0101, .0034)	(−.0046, .0063)	(−.0017, .0049)
Female	.1778***	−.0705***	−.0241***	−.0354***	−.0121***
	(.0769, .2786)	(−.0986, −.0424)	(−.0379, −.0104)	(−.0463, −.0244)	(−.0191, −.0051)
Grade	−.0027	.0429	.0125	−.0084	−.0165*
	(−.2128, .2075)	(−.0119, .0976)	(−.0132, .0382)	(−.0309, .0142)	(−.0313, −.0017)
Black	.6269***	.1045***	.0196	.0383**	.0367***
	(.3888, .8651)	(.0435, .1654)	(−.0102, .0495)	(.0138, .0628)	(.0199, .0535)
Hispanic	.3362***	.1137***	.0638***	.0211*	−.0063
	(.1664, .5061)	(.0678, .1596)	(.0413, .0863)	(.0033, .0390)	(−.0176, .0050)
Asian	−.1099	−.0719*	−.0077	.0074	.0050
	(−.3156, .0958)	(−.1326, −.0112)	(−.0356, .0202)	(−.0134, .0283)	(−.0076, .0175)
Other race/ethnicity	.5448***	.1238***	.0454**	.0102	.0206*
	(.2975, .7920)	(.0554, .1922)	(.0112, .0797)	(−.0153, .0356)	(.0043, .0370)
Household income	−.0034***	−.0007***	−.0004***	−.0003***	−.0001***
	(−.0043, −.0024)	(−.0010, −.0004)	(−.0005, −.0002)	(−.0004, −.0002)	(−.0002, −.0001)
Residents in household	−.0261	−.0005	−.0024	−.0021	−.0003
	(−.0656, .0134)	(−.0113, .0103)	(−.0076, .0029)	(−.0063, .0022)	(−.0030, .0024)
Parents’ education	−.0869***	−.0220***	−.0098***	−.0054***	−.0019*
	(−.1109, −.0630)	(−.0287, −.0152)	(−.0129, −.0066)	(−.0079, −.0029)	(−.0035, −.0003)
BMI in fall of kindergarten	1.5225***	.3098***	.1098***	.1018***	.0629***
	(1.4843, 1.5607)	(.2990, .3206)	(.1056, .1141)	(.0985, .1052)	(.0602, .0656)
Age in fall of kindergarten	−.0188	−.0041	.0002	−.0028	−.0027
	(−.0718, .0343)	(−.0192, .0109)	(−.0066, .0070)	(−.0083, .0027)	(−.0060, .0005)
2011 cohort	−.2030*	−.0742**	−.0087	−.0054	−.0051
	(−.3730, −.0330)	(−.1201, −.0284)	(−.0301, .0126)	(−.0219, .0110)	(−.0151, .0049)
Students	13,920	13,920	13,920	13,920	13,920

Open in a separate window

Abbreviations: BMI, body mass index; PA, physical activity; PE, physical education.

***p < 0.001, **p < 0.01, *p < 0.05. School‐clustered 95% confidence intervals are in parentheses. State policy variables are in bold. All models include state and cohort fixed effects. Data sources are the same as in Table2.

Limitations

Our analysis does have limitations. Although BMI was carefully measured, time in PE and recess was measured coarsely and inconsistently across different cohorts, making it challenging in some cases to tell whether a school had complied with state laws11 or whether PE and recess increased or decreased between the older and younger cohorts. We addressed the coarse measurement of time by using interval regression, but our interval regression model assumed that residuals are normal and hom*oscedastic, and departures from those assumptions might have introduced bias.23

This was an observational study, and it is possible that changes in BMI, PE time, or recess time were influenced by changes in some confounding variable that was correlated with changes in state policies. We reduced the danger of confounding by using a difference‐in‐differences design with state fixed effects, cohort fixed effects, and controls for initial BMI and other covariates. In addition, supplemental analyses showed that our results did not change when further covariates were added to the model. Nevertheless, it is always possible that missing covariates might have confounded the estimates. Possible omitted variables include changes in diet, increasing use of attention‐deficit/hyperactivity disorder medications, and changes in the built environment between the earlier cohort and the later cohort.24, 25

That said, concerns about confounding and bias are more common when a policy appears effective than when it does not. It is relatively easy to imagine how a confounder could make an ineffective policy appear effective but harder to imagine how a confounder could make an effective policy appear ineffective. To produce an estimated effect of zero, a confounder would have to have an effect that was exactly equal and opposite to the policy's effect—which is not impossible, but it seems a bit contrived.

It is also important to remember that we found no effect of state PE and PA laws on the time that children actually spent in PE or recess. In light of that finding, it is easier to understand why the laws had no effect on BMI. It is not that the PE/PA “treatment” was ineffective; instead, the treatment was not actually applied.

Discussion

Between 1998 and 2016, half of US states passed laws meant to increase the time that elementary students spent in school PE or PA. Two‐thirds of those states now require or at least recommend PE or PA for the 150 minutes/week that is endorsed by medical authorities. Yet, relative to states that did not change their laws, states that increased PE or PA requirements failed to reduce children's BMIs, overweight prevalence, or obesity prevalence. Changes to laws even failed to increase the time that children actually spent in PE or recess.

Weight control is hardly the only reason for children to be physically active. Regular, moderate to vigorous PA has a variety of physical, emotional, cognitive, and social benefits.26, 27 Yet it seems unlikely that children will reap these benefits if schools fail to comply with state laws for PE and PA time.

Although past studies also found that schools often fail to comply with state laws for PE time,10, 11, 28 the finding that changes to state law failed to cause any increase in the time spent in PE or recess is a bit startling. There are several possible explanations. Some schools may have been unaware of changes in state requirements. Some schools might have had trouble finding time in their schedules, and some states may not have provided clear guidance about what other activities could be cut to make more time for PA. PA may appear to compete with other priorities—especially reading and math scores—for which accountability is stronger,29 yet systematic evidence reviews suggest that reallocating modest amounts of time to PE from other activities does not reduce academic achievement, and may even increase achievement in some cases.30, 31, 32

There are several potential ways to improve schools’ compliance with state laws governing PE and PA time. States might consider providing model schedules, or at least providing clearer guidance regarding what activities can be cut to make more time for PE or recess. States might require schools to submit their schedules, so that state officials can verify that schools are offering enough PE and PA time. This may get easier as more schools adopt scheduling software that lets them report schedules in a standard format.

If states fail to enforce their own laws for school PE and PA, one option is for parents and advocates to sue school districts for failing to comply. In California, four lawsuits led targeted districts to increase PE time and improve reporting.33 In the affected districts, students’ cardiovascular fitness improved,28 though effects on obesity were not estimated.

If schools complied with state laws, could laws requiring 150 minutes/week of PE curb the obesity epidemic? Our data cannot answer that question, but a back‐of‐the‐envelope calculation suggests any effect of PE on obesity would be very small. Consider a 2003 study of third graders in ten cities,34 which found that students spent an average of 69 minutes/week in PE. Out of that time, students spent only 10 minutes/week being very active, versus 15 minutes/week walking, 25 minutes/week standing (for example, standing in line), 17 minutes/week sitting (for example, watching a demonstration or sitting on a courtwide bench), and 1 minute/week lying down In total, this weekly mix of activities burned 5 kcal/kg/week, or 165 kcal/week for an 8‐year‐old of average weight (33 kg)—just 60 kcal/week more than the child would have expended if they spent the same 69 minutes at their desks.34 At the same intensity level, increasing PE from 69 to 150 minutes/week would have increased energy expenditure by an additional 70 kcal/week [60 × (150 − 69)/69]. An increase in expenditure of 70 kcal/week is just 4%‐5% of the energy imbalance that is thought to have caused the epidemic of child obesity—an energy imbalance that has been estimated at 1,666 kcal/week for 8‐year‐old boys, or 1,386 kcal/week for 8‐year‐old girls.35

This calculation suggests that at typical intensity levels, laws increasing PE to 150 minutes/week would not be enough to noticeably reduce child obesity, even if compliance were perfect. To increase the potential impact of PE, schools should increase intensity—for example, by adopting PE curricula in which children spend more time being active and less time sitting or standing. PE intensity is challenging for regulators or even PE teachers to track, but it may become easier as motion‐sensing fitness‐tracking devices become more common. Yet even a vigorous PE program may have limited effects; RCT evidence suggests that even carefully designed PA interventions often fail to reduce BMI in school‐age children.7

School PA policies, by themselves, will probably not be enough to curb the obesity epidemic. The obesity epidemic was caused primarily by an increase in consumption, rather than a reduction in PA,36, 37 and antiobesity interventions are far more effective when they reduce consumption than when they increase PA alone.7, 38 Yet the obesity effects of school nutrition policies can be disappointing as well. For example, school soda bans have had little impact on obesity,39 perhaps because students and families found other places, outside of school, to purchase sugar‐sweetened beverages.40

Both PA and dietary interventions may have little effect if they are implemented only at school. Even on school days, children get only one‐fifth of their daily calories from the school cafeteria,41 and children gain weight faster when school is out than when schools is in.42, 43 To reverse the obesity epidemic, interventions must address consumption and activity not just in school, but outside of school as well.

Funding/Support: We thank the Policy Research Institute at the University of Texas, Austin, for a $15,000 grant supporting early work on this project.

Conflict of Interest Disclosures: The authors have no conflicts to disclose.

Acknowledgments: We thank Marilyn Headley for updating the data on state laws concerning physical education and other forms of school‐based physical activity.

Table Appendix 1

This table starts with the model predicting fifth grade BMI. It then shows that the policy effects estimated from that model do not change when we add three new covariates: after‐school care, an indicator for whether a school qualified for Title I funds, and the percentage of students at each school who qualified for free or reduced‐price lunch. We also found that results changed little when we predicted zBMI, overweight, or obesity

Predictors	Base model (from Table4)	Add after‐school care	Add school % free and reduced lunch	Add title I school
PE min/week required	−.0006	−.0002	−.0007	.0004
	(−.0035,.0023)	(−.0032,.0027)	(−.0038,.0024)	(−.0027,.0035)
PE min/week recommended	−.0020	−.0014	−.0023	−.0016
PE min/week recommended	(−.0056,.0017)	(−.0051,.0022)	(−.0061,.0015)	(−.0055,.0022)
PA min/week required	−.0003	−.0003	−.0013	.0005
	(−.0026,.0020)	(−.0026,.0020)	(−.0038,.0012)	(−.0020,.0029)
Age	.0472	.0458	.0401	.0586*
	(−.0067,.1011)	(−.0092,.1009)	(−.0189,.0992)	(.0019,.1154)
Female	.1778***	.1816***	.2144***	.1993***
	(.0769,.2786)	(.0790,.2841)	(.1034,.3255)	(.0933,.3052)
Grade	−.0027	.0162	−.0065	−.0299
	(−.2128,.2075)	(−.1954,.2278)	(−.2332,.2202)	(−.2486,.1888)
Black	.6269***	.6535***	.4868***	.5559***
	(.3888,.8651)	(.4104,.8967)	(.2216,.7521)	(.3013,.8104)
Hispanic	.3362***	.3439***	.2169*	.3001**
	(.1664,.5061)	(.1731,.5146)	(.0236,.4103)	(.1168,.4833)
Asian	−.1099	−.0749	−.2416*	−.1213
	(−.3156,.0958)	(−.2871,.1372)	(−.4665,−.0167)	(−.3429,.1003)
Other race/ethnicity	.5448***	.5430***	.4783***	.5487***
	(.2975,.7920)	(.2876,.7983)	(.2169,.7397)	(.2899,.8075)
Household income	−.0034***	−.0034***	−.0029***	−.0032***
	(−.0043,−.0024)	(−.0044,−.0024)	(−.0041,−.0018)	(−.0043,−.0022)
Residents in household	−.0261	−.0227	−.0256	−.0301
	(−.0656,.0134)	(−.0632,.0177)	(−.0691,.0179)	(−.0721,.0120)
Parents’ education	−.0869***	−.0851***	−.0723***	−.0767***
	(−.1109,−.0630)	(−.1097,−.0604)	(−.1000,−.0446)	(−.1023,−.0512)
BMI (fall K)	1.5225***	1.5321***	1.5210***	1.5251***
	(1.4843,1.5607)	(1.4928,1.5715)	(1.4781,1.5638)	(1.4844,1.5658)
Age (fall K)	−.0188	−.0160	−.0115	−.0268
	(−.0718,.0343)	(−.0702,.0383)	(−.0695,.0465)	(−.0828,.0292)
2011 Cohort	−.2030*	−.2235*	−.2327*	−.2774**
	(−.3730,−.0330)	(−.3955,−.0515)	(−.4265,−.0389)	(−.4609,−.0939)
After‐school care		−.1121
		(−.2622,.0380)
School % free/reduced lunch			.0043**
			(.0015,.0071)
Title I school				.2941***
				(.1653,.4229)
Number of children	13,920	13,530	11,740	12,670

Open in a separate window

***p < 0.001, **p < 0.01, *p < 0.05. 95% confidence intervals in parentheses. State policy variables are in bold. All models include state and cohort fixed effects.

variables are in bold. All models include state and cohort fixed effects.

Table Appendix 2

Models by sex

Panel A: Boys
	BMI	BMI Z score	Overweight	Obesity	Extreme obesity
PE min/week required	−.0002	.0005	−.0000	−.0000	.0001
	(−.0044, .0040)	(−.0007, .0016)	(−.0006, .0005)	(−.0005, .0005)	(−.0002 − .0004)
PE min/week recommended	−.0035	−.0004	−.0004	−.0003	−.0002
	(−.0075, .0005)	(−.0015, .0008)	(−.0010, .0001)	(−.0008, .0001)	(−.0004 −.0001)
PA min/week required	−.0015	−.0001	−.0004	−.0001	.0000
	(−.0047, .0016)	(−.0009, .0008)	(−.0008, .0001)	(−.0005, .0002)	(−.0002 −.0002)
Boys	7,050	7,050	7,050	7,050	7,050

Open in a separate window

Panel A: Girls
	BMI	BMI Z score	Overweight	Obesity	Extreme obesity
PE min/week required	−.0015	−.0010	−.0008**	.0001	.0001
	(−.0055, .0024)	(−.0021, .0001)	(−.0013, −.0002)	(−.0003, .0006)	(−.0002 −.0004)
PE min/week recommended	−.0002	−.0003	.0001	−.0002	.0001
	(−.0061, .0057)	(−.0017, .0012)	(−.0007, .0008)	(−.0007, .0003)	(−.0002 −.0003)
PA min/week required	.0012	.0004	.0002	.0001	−.0000
	(−.0018, .0042)	(−.0004, .0012)	(−.0002, .0006)	(−.0002, .0004)	(−.0002 −.0002)
Girls	6,870	6,870	6,870	6,870	6,870

Open in a separate window

*** p < 0.001, ** p < 0.01, * p < 0.05

Notes: Models are the same as in Table4, but here the data are separated by sex and the covariates are not shown. 95% confidence intervals are in parentheses.

Table Appendix 3

State Physical Education Requirements for Grades K‐8 in 1998–1999 through 2015–2016

State	Requirements
Alabama	Physical Education Policy: Beginning at least as far back as 1940, per Code of Alabama § 16‐40‐1, schools are required to offer physical education. Beginning in 2003–2004, all students in grades K‐8 are required to participate in daily physical education, with no exceptions or substitutions, and at least 30 minutes each day (150 minutes per week) are required for students in grades K‐6. At least 50 minutes each day are recommended by the Department of Education for students in grades 7–8. Applicable statutes and sources: Code of Alabama § 16‐40‐1; Code of Alabama, 1975, §16‐35‐4; Code of Ala. § 16‐35‐5; Alabama Course of Study Physical Education(2009).
Alaska	Physical Education Policy: 1998, per Alaska Stat. § 14.30.360, encourages each school district to implement a K‐12 program in health education, which should include instruction in physical education. Effective October 16, 2016, section c was added to Alaska Stat. § 14.30.360, which requires schools districts to establish guidelines for schools to provide opportunities for students in grades K‐8 to receive 90 percent of the daily amount of physical activity recommended for children and adolescents in the physical activity guidelines by the Centers for Disease Control and Prevention, United States Department of Health and Human Services. Applicable statutes and sources: Alaska Stat. § 14.30.360; Alaska. Administrative Code: 04 AAC 06.075; AS 14.07.060.
Arizona	Physical Education Policy: Beginning in 1993, per statute A.A.C. § R7‐2‐301, students in grades K – 8 are required to demonstrate a minimum level of competency in physical education. There is no requirement to participate in physical education or for schools to offer physical education. Applicable statutes and sources: A.A.C. § R7‐2‐301.
Arkansas	Physical Education Policy: Since 2001, per A.C.A. § 6‐16‐132; 2001 Ark. ALS 1748; students in grades K‐9 have been required to participate in physical education for no less than 1 hour per week which includes no less than 20 minutes of physical activity three times a week. │ 2003, per statute A.C.A. § 6‐16‐132;Ark. ALS 1729, every kindergarten through grade eight (K‐8) public educational institution in this state shall require no less than a total of one (1) hour per week of physical education training and instruction for every student who is physically fit and able to participate.│ 2007, per A.C.A. § 6‐16‐132; Ark. ALS 317; Arkansas Code Section 6‐16‐132(b)(1), requires every public‐school student, grades K – 6, who is physically fit and able to participate in sixty (60) minutes of physical education training and instruction each calendar week of the school year, which may include without limitation daily recess, physical education instruction in addition to the requirements of subdivision (B)(1)(A)(I)(A) of this section, or intramural sports. And ninety (90) minutes of physical activity each calendar week of the school year, and in grade 5 – 8 at schools organized to teach those grades. │Approved April 2015, per Ark. Acts 1079; SECTION 1. Arkansas Code Section 6‐16‐130(a); A.C.A. § 6‐16‐132; Grade K – 6, at least forty (40) minutes as determined by the superintendent of the school district of physical education training and instruction each calendar week of the school year; and (b) Ninety (90) minutes of physical activity each calendar week of the school year, which may include without limitation daily recess, physical education instruction in addition to the requirement of subdivision (b)(1)(A)(i)(a) of this section, or intramural sports; (ii) For students in grades five through eight (5‐8) who attend a public school organized to teach grades five through eight (5‐8), or any combination thereof, at least forty (40) minutes as determined by the superintendent of the school district, of physical education training and instruction each calendar week of the school year or an equivalent amount of time in each school year, with no additional requirement for physical activity.
California	Physical Education Policy: Enacted 1976, operative 1977, per Cal Ed Code § 51210, students in grades 1 to 6, inclusive, have been required to participate in physical education for a period of not less than 200 minutes each 10 schooldays, exclusive of recesses and the lunch period, and pupils in grades 7 to 12, inclusive, have been required to participate in physical education for a period of not less than 400 minutes each 10 schooldays. Recess may not be included in these requirements. Applicable statutes and sources: Cal Ed Code § 51210, 51222, 51220, 51223; Cal Ed Code § 33350; Cal Ed Code § 33352.
Colorado	Physical Education Policy: Effective 2008, per C.R.S. 22‐7‐1005, on or before December 15, 2009, the state board shall adopt standards that identify the knowledge and skills that a student should acquire as the student progresses from preschool through elementary and secondary education. Section (2)(a) The state board shall ensure that the preschool through elementary and secondary education standards, at a minimum, include standards in … physical education. Applicable statutes and sources: C.R.S. 22‐7‐1005. Physical Activity Policy: Effective April 2011 – 2012 school year, per Colo. Rev. Stat. § 22‐32‐136.5, school district board of education shall establish a monthly minimum requirement for physical activity opportunities for elementary students of at least 600 minutes per month (30 minutes per day) for schools that meet 5 days per week and students attend for a full school day (alternate time requirements are provided for schools that do not meet 5 days per week or enroll full‐day students). This requirement may be satisfied by recess, physical education class, fitness breaks, classroom activities that include physical activity, exercise programs, or field trips that include physical activity. Applicable statutes and sources: Colo. Rev. Stat. § 22‐32‐136.5; C.R.S. 22‐7‐1005.
Connecticut	Physical Education Policy: 1997, per Conn. Gen. Stat. § 10–16b, students in public schools are required to receive physical education instruction. Applicable statutes and sources: Conn. Gen. Stat. § 10–16b, Conn. Gen. Stat. § 10–221a. Physical Activity Policy: Beginning in 2004–2005, schools are required to offer a daily period of physical exercise to students in grades K‐5. Effective 2012, per Conn. Gen. Stat. § 10–221o, the state requires elementary schools to provide at least 20 minutes of daily physical exercise for K‐5 students. Applicable statutes and sources: Substitute House Bill No. 6525 (2013); Conn. Gen. Stat. § 10–221o.
D.C.	Physical Education Policy: Since 2010, per D.C. Code § 38–824.02, public schools and public charter schools shall provide physical education as follows: (1) For students in Kindergarten through Grade 5: (A) School years 2010–2011 to 2013–2014: an average of at least 30 minutes per week or the same level of physical education as provided in school year 2009–2010, whichever is greater; and (B) School year 2014–2015 and after: an average of at least 150 minutes per week; (2) For students in Grades 6 through 8: (A) School years 2010–2011 to 2013–2014: an average of at least 45 minutes per week or the same level of physical education as provided in school year 2009–2010, whichever is greater; and (B) School year 2014–2015 and after: an average of at least 225 minutes per week. (3) At least 50% of physical education class time shall be devoted to actual physical activity, with as much class time as possible spent in moderate‐to‐vigorous physical activity. Applicable statutes and sources: D.C. Code § 38–824.02
Delaware	Physical Education Policy: 2005, per Administrative Code 14:503, all public‐school students in each grade 1 to 8 shall be enrolled in a physical education program. 2006, per CDR 14–500, local school districts and charter schools shall provide instructional programs in physical education for each grade K to 12. Applicable statutes and sources: CDR 14–500; 14 Del. Admin. Code § 503; 14 DE Admin. Code 501.
Florida	Physical Education Policy: For 2006–2007, districts were encouraged to provide 150 minutes per week for students in grades K‐5 and 225 minutes per week for students in grades 6–8. Effective May 2007, per Fla. Stat. § 1003.455, each district school board shall provide 150 minutes of physical education each week for students in kindergarten through grade 5 and for students in grade 6 who are enrolled in a school that contains one or more elementary grades so that on any day during which physical education instruction is conducted there are at least 30 consecutive minutes per day. Applicable statutes and sources: 2006 Fl. ALS 301, Fla. Stat. § 1003.455, Fla. Stat. § 1003.4282.
Georgia	Physical Education Policy: Effective 1990, per O.C.G.A. § 20‐2‐142, the State Board of Education shall prescribe a course of study in health and physical education for all grades and grade levels in the public schools and shall establish minimum time requirements and standards for its administration. Effective 2000, per Ga. Comp. R. & Regs. r. 160‐4‐2‐.12, (section b) Each school containing any grade K‐5 shall provide a minimum of 90 contact hours of instruction at each grade level K‐5 in health and physical education. (c) Each school containing any grade 6–12 shall make available instruction in health and physical education. Applicable statutes and sources: O.C.G.A. § 20‐2‐142; Ga. Comp. R. & Regs. r. 160‐4‐2‐.12; Ga. Comp. R. & Regs. r. 160‐4‐2‐.47.
Hawaii	Physical Education Policy: Physical education requirements were phased in over a four year period through school year 2010–2011. Per the state wellness requirements, schools are required to offer 45 minutesof physical education classes per week for grades K‐3, 55 minutes per week for grades 4–5, 107 minutes per week for elementary grade 6, and200 minutes per week for secondary grades 6–12. Applicable statutes and sources: Policy 1110–6 (1995), amended: 08/17/2006, and renamed POLICY 103‐1 in 2016, State Department of Education Content and Performance Standards for physical education. Physical Activity Policy: Beginning in 2010–2011, students in all grades are required to have at least 20 minutes a day of supervised recess. Applicable statutes and sources: State Department of Education Content and Performance Standards for physical education.
Idaho	Physical Education Policy: Effective in 2006, elementary and middle schools are required to offer physical education. Applicable statutes and sources: IDAPA 08.02.03.104, State Board of Education Administrative Rules 08.02.02.104.01‐04. Physical Activity Policy: Beginning in 1963, schools are required to offer physical fitness. Applicable statutes and sources: Idaho Code § 33–1605.
Illinois	Physical Education Policy: Since 1996, per 105 ILCS 5/27‐5 and 23 Ill. Adm. Code 1.420, all students are required to engage in daily physical education. Applicable statutes and sources: 105 ILCS 5/27‐6; 105 ILCS 5/27‐5; 23 Ill. Adm. Code 1.420.
Indiana	Physical Education Policy: Beginning in 2010, per 511 IAC 6.1‐5‐2.6, the elementary school curriculum is required to include physical education. Beginning in 2010, per 511 IAC 6.1‐5‐3.6, the middle school curriculum is required to include physical education. The Indiana State Board of Education recommends that students in grades 1–3 participate in 105 minutes per week of motor skills development and health/safety education, students in grades 4–6 participate in 75 minutes of PE per week, and students in grades 6–8, when grade 6 is included in a middle school, participate in 100 minutes of PE per week. Applicable statutes and sources: 511 IAC 6.1‐5‐2.5; 511 IAC 6–7.1‐5; 511 IAC 6.1‐5‐3.6, Indiana State Board of Education. Physical Activity Policy: Beginning in 2005–2006, per Burns Ind. Code Ann. § 20‐30‐5‐7, each school shall include physical fitness as part of the curriculum. Beginning in 2006–2007, per Burns Ind. Code Ann. § 20‐30‐5‐7.5, elementary schools are required to provide daily physical activity, which may include recess. Applicable statutes and sources: Burns Ind. Code Ann. § 20‐30‐5‐7, Burns Ind. Code Ann. § 20‐30‐5‐7.5.
Iowa	Physical Education Policy: Beginning in 1986, per Iowa Code § 256.11, physical education is required to be taught in grades 1 – 8. Applicable statute or sources: Iowa Code § 256.11. Physical Activity Policy: Beginning in 2009, per Iowa Code § 256.11, students in grades K‐5 are required to engage in 150 minutes per week (30 minutes per day) of physical activity and students in grades 6–12 are required to engage in 120 minutes per week of physical activity. Applicable statutes and sources: Iowa Code § 256.11.
Kansas	Physical Education Policy: Elementary and secondary schools, to receive accreditation, are required to offer physical education. Applicable statutes and sources: K.A.R. § 91‐31‐32; K.A.R. § 91‐31‐35. The available state statutes date back only to 2005. According to the Shape of the Nation and Trust for America's Health reports, the requirement dates back to at least 1993.
Kentucky	Physical Education Policy: Beginning in 2006–2007, students in grades K‐8 are required to participate in PE as part of the state curriculum. Applicable statute and sources: 704 KAR 3:303; 704 KAR 3:305.
Louisiana	Physical Education Policy: Since 1980, per La. R.S. § 17:276, LA secondary schools have been required to offer physical education courses. Applicable statutes and sources: La. R.S. § 17:276. Physical Activity Policy: Beginning in 2004–2005, per 2004 La. ALS 734, schools that include grades K‐6 are required to provide 150 minutes of moderate‐to‐vigorous physical activity for students. In 2009, per La. R.S. § 17:17.1, schools that include grades K‐8 are required to provide 150 minutes of moderate‐to‐vigorous physical activity for students. Applicable statutes and sources: 2004 La. ALS 734, 2009 La. ALS 286, La. R.S. § 17:17.1, State Bulletins 102 and 103.
Maine	Physical Education Policy: Since 1983, elementary and secondary schools shall offer physical education. Since 2009, elementary schools must offer physical education as part of the course of study. Effective January 1985, per CMR 94‐348‐004,physical education was required in secondary schools. Effective 2007, per CMR 05‐071‐132 and 20‐A M.R.S.A. § 1454, requires physical education instruction for prek – 8. Since 2009, per 2009 ch. 313, § 11 (AMD) and 20‐A M.R.S. § 4711,elementary, middle and junior high (per 20‐A M.R.S. § 4712) and secondary schools (per 20‐A M.R.S. § 4721)schools must provide for the instruction of all students….physical education. Applicable statutes and sources: 20‐A M.R.S. § 4711; 20‐A M.R.S. § 4723; 2001 Me. ALS 454.
Maryland	Physical Education Policy: Since 1996, per Md. EDUCATION Code Ann. § 7–409, each school shall provide physical education to all students in grades K – 12. Applicable statutes and sources: Md. EDUCATION Code Ann. § 7–409.
Massachusetts	Physical Education Policy: Since 1962, per ALM GL ch. 71, § 1, physical education shall be taught as a required subject in all grades for all students in the public. Applicable statutes and sources: ALM GL ch. 71, § 1; ALM GL ch. 71, § 3.
Michigan	Physical Education Policy: Since 1976, per MCLS § 380.1502, physical education shall be provided to all students in all public schools. Beginning in 2003–2004, the state Board of Education recommends that all schools offer PE for 150 minutes per week for grades K‐5 and for 225 minutes per week for grades 6–8. Applicable statutes and sources: MCLS § 380.1502; MCLS § 380.1278a.
Minnesota	Physical Education Policy: Beginning in 2005–2006 per 2005 Minn. ALS 5, Statute 120A‐22, students in grades 1–10 are required to participate in physical education. Beginning in the 2016–2017 school year, per Minn. Stat. § 120B.021, the department must adopt the most recent National Association of Sport and Physical Education standards and benchmarks for grades K‐12. Applicable statutes and sources: 2005 Minn. ALS 5, Statute 120A‐22; Minn. Stat. § 120B.021.
Mississippi	Physical Education Policy: Between 2002–2003 and 2006–2007, the legislature recommends that students in grades K‐8 participate in PE for 150 minutes per week. Beginning in 2007–2008, per Miss. Code Ann. § 37‐13‐134, students in grades K‐8 are required to participate in physical activity instruction for 150 minutes per week. Applicable statutes: Miss. Code Ann. § 37‐13‐134; 2002 Miss. ALS 585; Miss. Code Ann. § 37‐11‐8.
Missouri	Physical Education Policy: Effective 1995, per § 161.102 R.S.Mo, schools are required to provide PE to all students in all public schools. Beginning in 2001–2002, per 5 CSR 20–100.255, students in grades K‐5 are required to participate in 50 minutes per week of PE and students in grades 6–8 are required to receive 3000 minutes of instruction in PE per year. Beginning with the school year 2010–11, per § 167.720 R.S.Mo., school districts shall ensure that students in elementary schools participate in moderate physical activity for the entire school year of an average of 150 minutes per five‐day school week, or an average of 30 minutes per day. Students in middle schools may, at the school's discretion, participate in at least 225 minutes of physical activity per school week. Applicable statutes: § 161.102 R.S.Mo. and Missouri School Improvement Program; § 167.720 R.S.Mo; 5 CSR 20–100.255; 200 MSGR 126.
Montana	Physical Education Policy: Since 1998, schools are required to offer PE. Applicable statutes and sources: MONT. ADMIN. R. 10.54.2501.
Nebraska	Physical Education Policy: Per Nebraska Admin. Code Title 92, Ch. 10 004.03A9, schools are required to offer physical education in elementary and middle schools. Applicable statutes and sources: Nebraska Admin. Code Title 92, Ch. 10‐004.03A9; Nebraska Admin. Code Title 92, Ch. 10 004.02A6.
Nevada	Physical Education Policy: Beginning in 1999–2000, PE “must be taught as applicable for grade levels and to the extent practicable in all public schools.” Applicable statutes and sources: Nev. Rev. Stat. Ann. § 389.018; CHAPTER: 389 LCB File No. R037‐12A.
New Hampshire	Physical Education Policy: Beginning in 1975 or possibly earlier, per RSA 189:10, PE is required to be taught to students in all grades. Beginning in 2004–2005, per RSA 189:11‐a, the school boards of each school district were required to adopt a policy recommending daily PA. Effective March 2014, per 231 NH GOV REG 24, NH readopted Ed 306.41 to require physical education in k‐ 12. Applicable statutes and sources: RSA 189:10; RSA 189:11‐a; 231 NH GOV REG 24.
New Jersey	Physical Education Policy: 1967, per N.J. Stat. § 18A:35‐5 and N.J. Stat. § 18A:35‐8, students are required to take courses in PE and health in grades 1–12 for a combination of 150 minutes per week. Local school districts determine the percentage of the 150 minutes that is allocated to PE. Applicable statutes and sources: N.J. Stat. § 18A:35‐7; N.J. Stat. § 18A:35‐8; N.J. Stat. § 18A:35‐5.
New Mexico	Physical Education Policy: Beginning in 2005–2006, schools are required to provide PE in grades 1–8. Prior to 2005, schools are required to provide physical fitness to grades 4–8. Applicable statutes and sources: N.M. Stat. Ann. § 22‐13‐1.
New York	Physical Education Policy: Effective 1983, per 8 NYCRR § 135.4, students in grades K‐3 must participate in PE daily and students in grades 4–6 must participate in PE at least 3 times per week. Students in grades K‐6 must participate in at least 120 minutes per week. Schools are required to offer PE to students in grades 7–8 at least 3 times per week in one semester and 2 times per week in the other semester. Applicable statutes and sources: 8 NYCRR § 135.4.
North Carolina	Physical Education Policy: Since 1995 per N.C. Gen. Stat. § 115C‐12, PE shall be offered to every student in the state. By the 2006 school year Policy ID Number:GCS‐S‐000 shall be fully implemented. Per Policy ID Number:GCS‐S‐000, the recommended minimum amount of PE is 150 minutes per week for students in grades K‐5 and 225 minutes per week for students in grades 6–8. Applicable statutes and sources: Policy ID Number:GCS‐S‐000, N.C. Gen. Stat. § 115C‐12.
North Dakota	Physical Education Policy: Since 2001, public and nonpublic elementary and middle schools (grades K – 8) shall provide students instruction in physical education. Applicable statutes and sources: N.D. Cent. Code, § 15.1‐21‐01.
Ohio	Physical Education Policy: Since 1993, schools are required to offer PE for students in all grades. Applicable statutes and sources: OAC Ann. 3301‐35‐04; ORC Ann. 3313.60, 1991 Ohio SB 195.
Oklahoma	Physical Education Policy: Beginning in 2006–2007, schools are required to provide PE for elementary school students for at least 60 minutes per week. Applicable statutes and sources: 70 Okl. St. § 11–103.9.
Oregon	Physical Education Policy: Beginning in 1997–1998, per Or. Admin. R. 581‐022‐1210, schools are required to offer PE to students in all grades. Effective the 2017–2018 school year, per ORS § 329.496 every public school student in kindergarten through grade 8 shall participate in physical education for the entire school year. Students in kindergarten through grade 5 shall participate in physical education for at least 150 minutes during each school week. Students in grades 6 through 8 shall participate in physical education for at least 225 minutes during each school week. Applicable statutes and sources: Or. Admin. R. 581‐022‐1210; ORS 329.045; ORS § 329.496.
Pennsylvania	Physical Education Policy: PE is required to be taught to students in all grades. Applicable statutes and sources: 22 Pa. Code § 4.27; 022 PA Code 4.21
Rhode Island	Physical Education Policy: Students are required to participate in PE for at least 20 minutes each day in grades 1–12. Recess cannot count towards the PE requirements. Applicable statutes and sources: Rules and Regulations of School Health Programs; R.I. Gen. Laws § 16‐22‐4
South Carolina	Physical Education Policy: Beginning in 2006–07, per S.C. Code Ann. § 59‐10‐10, students in grades K‐5 must be provided a minimum of 150 minutes a week of physical education and physical activity. In 2006–07, a minimum of 60 minutes a week must be provided in physical education, and as Section 59‐10‐20 is phased in, the minimum time for physical education must be increased to 90 minutes a week. Applicable statutes and sources: S.C. Code Ann. § 59‐10‐10; S.C. Code Ann. § 59‐10‐10.
South Dakota	Physical Education Policy: Schools are not required to provide PE to students in grades K‐8. Applicable statutes and sources: N/A
Tennessee	Physical Education Policy: Per Tenn. Comp. R. & Regs. R. 0520‐01‐03‐.05., schools are required to offer PE to students in grades K‐8. Beginning in 2006–2007, at least 90 minutes of physical activity must be provided throughout the day for students in grades K‐8. Applicable statutes and sources: Tenn. Comp. R. & Regs. R. 0520‐01‐03‐.05; T.C.A. § 49‐6‐1021.
Texas	Physical Education Policy: Schools must offer PE to students in grades K‐5 since at least 1993 per 19 TAC § 74.2. Beginning in 2002–2003, per Senate Bill 19 passed in 2001, students in grades K‐6 must participate in at least 30 minutes per day or 135 minutes per week of physical activity (PA). In 2004–2005, schools are required to offer PE to students in grades 6–8. Beginning in 2006–2007, per Senate Bill 42 passed in 2005, students in grades 6–8 are required to participate in PE for at least 30 minutes per day or 135 minutes per week or 225 minutes per fortnight. The requirements for grades K‐5 refer to physical activity and permit recess to contribute to this requirement. Since December 2009 per 19 TAC § 74.37, On a weekly basis, at least 50% of a physical education class shall be used for actual student physical activity and the activity shall be, to the extent practicable, at a moderate or vigorous level. Applicable statutes and sources: Administrative Codes 74.2, 74.32, and 74.3, Tex. Educ. Code § 28.002; 19 TAC § 74.37.
Utah	Physical Education Policy: Students in grades K‐6 are required to receive instruction in PE. Applicable statutes and sources: U.A.C. R277‐700‐4.
Vermont	Physical Education Policy: Schools are required to offer PE. The Nutrition and Physical Fitness Guidelines, established in 2005, recommend 150 minutes per week for elementary school students and 225 minutes per week for middle school students. Applicable statutes and sources: 16 V.S.A. § 906.
Virginia	Physical Education Policy: Beginning in 1999–2000, PE must be taught as part of the curriculum in all grades. Beginning in 2008–2009, a program of physical fitness, which can include PE, must be available to all students for at least 150 minutes per week. Applicable statutes and sources: Va. Code Ann. § 22.1‐253.13:1; Code 22.1‐207; Code 22.1‐200.
Washington	Physical Education Policy: Students in grades 1–8 shall receive instruction in PE. By 2010, per Rev. Code Wash. (ARCW) § 28A.210.365, all students in grades one through eight should have at least one hundred fifty minutes of quality physical education every week Applicable statutes and sources: Rev. Code Wash. (ARCW) § 28A.230.040; Rev. Code Wash. (ARCW) § 28A.210.365.
West Virginia	Physical Education Policy: Elementary and middle schools are required to provide PE. Beginning in 2005–2006, per § 18‐2‐7a., students in grades K‐5 are required to participate in at least 90 minutes of PE per week and students in grades 6–8 are required to participate in daily PE for one full period for one semester. Applicable statutes and sources: W. Va. Code § 18‐2‐7a
Wisconsin	Physical Education Policy: Beginning in 1988–1989, students in grades K‐6 are required to participate in PE at least 3 days per week and students in middle school are required to participate in PE at least once per week. Applicable statutes and sources: Wis. Stat. § 121.02, 1987 Wis. ALS 27.
Wyoming	Physical Education Policy: Beginning in 1998–1999, each school district shall provide educational programs that include PE. Students in grades 1–8 are required to be proficient in PE. Applicable statutes and sources: Wyo. Stat. § 21‐9‐101

State

Requirements

Alabama

Physical Education Policy: Beginning at least as far back as 1940, per Code of Alabama § 16‐40‐1, schools are required to offer physical education. Beginning in 2003–2004, all students in grades K‐8 are required to participate in daily physical education, with no exceptions or substitutions, and at least 30 minutes each day (150 minutes per week) are required for students in grades K‐6. At least 50 minutes each day are recommended by the Department of Education for students in grades 7–8.

Applicable statutes and sources: Code of Alabama § 16‐40‐1; Code of Alabama, 1975, §16‐35‐4; Code of Ala. § 16‐35‐5; Alabama Course of Study Physical Education(2009).

Alaska

Physical Education Policy: 1998, per Alaska Stat. § 14.30.360, encourages each school district to implement a K‐12 program in health education, which should include instruction in physical education. Effective October 16, 2016, section c was added to Alaska Stat. § 14.30.360, which requires schools districts to establish guidelines for schools to provide opportunities for students in grades K‐8 to receive 90 percent of the daily amount of physical activity recommended for children and adolescents in the physical activity guidelines by the Centers for Disease Control and Prevention, United States Department of Health and Human Services.

Applicable statutes and sources: Alaska Stat. § 14.30.360; Alaska. Administrative Code: 04 AAC 06.075; AS 14.07.060.

Arizona

Physical Education Policy: Beginning in 1993, per statute A.A.C. § R7‐2‐301, students in grades K – 8 are required to demonstrate a minimum level of competency in physical education. There is no requirement to participate in physical education or for schools to offer physical education.

Applicable statutes and sources: A.A.C. § R7‐2‐301.

Arkansas

Physical Education Policy: Since 2001, per A.C.A. § 6‐16‐132; 2001 Ark. ALS 1748; students in grades K‐9 have been required to participate in physical education for no less than 1 hour per week which includes no less than 20 minutes of physical activity three times a week. │ 2003, per statute A.C.A. § 6‐16‐132;Ark. ALS 1729, every kindergarten through grade eight (K‐8) public educational institution in this state shall require no less than a total of one (1) hour per week of physical education training and instruction for every student who is physically fit and able to participate.│ 2007, per A.C.A. § 6‐16‐132; Ark. ALS 317; Arkansas Code Section 6‐16‐132(b)(1), requires every public‐school student, grades K – 6, who is physically fit and able to participate in sixty (60) minutes of physical education training and instruction each calendar week of the school year, which may include without limitation daily recess, physical education instruction in addition to the requirements of subdivision (B)(1)(A)(I)(A) of this section, or intramural sports. And ninety (90) minutes of physical activity each calendar week of the school year, and in grade 5 – 8 at schools organized to teach those grades. │Approved April 2015, per Ark. Acts 1079; SECTION 1. Arkansas Code Section 6‐16‐130(a); A.C.A. § 6‐16‐132; Grade K – 6, at least forty (40) minutes as determined by the superintendent of the school district of physical education training and instruction each calendar week of the school year; and (b) Ninety (90) minutes of physical activity each calendar week of the school year, which may include without limitation daily recess, physical education instruction in addition to the requirement of subdivision (b)(1)(A)(i)(a) of this section, or intramural sports; (ii) For students in grades five through eight (5‐8) who attend a public school organized to teach grades five through eight (5‐8), or any combination thereof, at least forty (40) minutes as determined by the superintendent of the school district, of physical education training and instruction each calendar week of the school year or an equivalent amount of time in each school year, with no additional requirement for physical activity.

California

Physical Education Policy: Enacted 1976, operative 1977, per Cal Ed Code § 51210, students in grades 1 to 6, inclusive, have been required to participate in physical education for a period of not less than 200 minutes each 10 schooldays, exclusive of recesses and the lunch period, and pupils in grades 7 to 12, inclusive, have been required to participate in physical education for a period of not less than 400 minutes each 10 schooldays. Recess may not be included in these requirements.

Applicable statutes and sources: Cal Ed Code § 51210, 51222, 51220, 51223; Cal Ed Code § 33350; Cal Ed Code § 33352.

Colorado

Physical Education Policy: Effective 2008, per C.R.S. 22‐7‐1005, on or before December 15, 2009, the state board shall adopt standards that identify the knowledge and skills that a student should acquire as the student progresses from preschool through elementary and secondary education. Section (2)(a) The state board shall ensure that the preschool through elementary and secondary education standards, at a minimum, include standards in … physical education.

Applicable statutes and sources: C.R.S. 22‐7‐1005.

Physical Activity Policy: Effective April 2011 – 2012 school year, per Colo. Rev. Stat. § 22‐32‐136.5, school district board of education shall establish a monthly minimum requirement for physical activity opportunities for elementary students of at least 600 minutes per month (30 minutes per day) for schools that meet 5 days per week and students attend for a full school day (alternate time requirements are provided for schools that do not meet 5 days per week or enroll full‐day students). This requirement may be satisfied by recess, physical education class, fitness breaks, classroom activities that include physical activity, exercise programs, or field trips that include physical activity.

Applicable statutes and sources: Colo. Rev. Stat. § 22‐32‐136.5; C.R.S. 22‐7‐1005.

Connecticut

Physical Education Policy: 1997, per Conn. Gen. Stat. § 10–16b, students in public schools are required to receive physical education instruction.

Applicable statutes and sources: Conn. Gen. Stat. § 10–16b, Conn. Gen. Stat. § 10–221a.

Physical Activity Policy: Beginning in 2004–2005, schools are required to offer a daily period of physical exercise to students in grades K‐5. Effective 2012, per Conn. Gen. Stat. § 10–221o, the state requires elementary schools to provide at least 20 minutes of daily physical exercise for K‐5 students.

Applicable statutes and sources: Substitute House Bill No. 6525 (2013); Conn. Gen. Stat. § 10–221o.

D.C.

Physical Education Policy: Since 2010, per D.C. Code § 38–824.02, public schools and public charter schools shall provide physical education as follows:

(1) For students in Kindergarten through Grade 5:

(A) School years 2010–2011 to 2013–2014: an average of at least 30 minutes per week or the same level of physical education as provided in school year 2009–2010, whichever is greater; and

(B) School year 2014–2015 and after: an average of at least 150 minutes per week;

(2) For students in Grades 6 through 8:

(A) School years 2010–2011 to 2013–2014: an average of at least 45 minutes per week or the same level of physical education as provided in school year 2009–2010, whichever is greater; and

(B) School year 2014–2015 and after: an average of at least 225 minutes per week.

(3) At least 50% of physical education class time shall be devoted to actual physical activity, with as much class time as possible spent in moderate‐to‐vigorous physical activity.

Applicable statutes and sources: D.C. Code § 38–824.02

Delaware

Physical Education Policy: 2005, per Administrative Code 14:503, all public‐school students in each grade 1 to 8 shall be enrolled in a physical education program. 2006, per CDR 14–500, local school districts and charter schools shall provide instructional programs in physical education for each grade K to 12.

Applicable statutes and sources: CDR 14–500; 14 Del. Admin. Code § 503; 14 DE Admin. Code 501.

Florida

Physical Education Policy: For 2006–2007, districts were encouraged to provide 150 minutes per week for students in grades K‐5 and 225 minutes per week for students in grades 6–8. Effective May 2007, per Fla. Stat. § 1003.455, each district school board shall provide 150 minutes of physical education each week for students in kindergarten through grade 5 and for students in grade 6 who are enrolled in a school that contains one or more elementary grades so that on any day during which physical education instruction is conducted there are at least 30 consecutive minutes per day.

Applicable statutes and sources: 2006 Fl. ALS 301, Fla. Stat. § 1003.455, Fla. Stat. § 1003.4282.

Georgia

Physical Education Policy: Effective 1990, per O.C.G.A. § 20‐2‐142, the State Board of Education shall prescribe a course of study in health and physical education for all grades and grade levels in the public schools and shall establish minimum time requirements and standards for its administration. Effective 2000, per Ga. Comp. R. & Regs. r. 160‐4‐2‐.12, (section b) Each school containing any grade K‐5 shall provide a minimum of 90 contact hours of instruction at each grade level K‐5 in health and physical education. (c) Each school containing any grade 6–12 shall make available instruction in health and physical education.

Applicable statutes and sources: O.C.G.A. § 20‐2‐142; Ga. Comp. R. & Regs. r. 160‐4‐2‐.12; Ga. Comp. R. & Regs. r. 160‐4‐2‐.47.

Hawaii

Physical Education Policy: Physical education requirements were phased in over a four year period through school year 2010–2011. Per the state wellness requirements, schools are required to offer 45 minutesof physical education classes per week for grades K‐3, 55 minutes per week for grades 4–5, 107 minutes per week for elementary grade 6, and200 minutes per week for secondary grades 6–12.

Applicable statutes and sources: Policy 1110–6 (1995), amended: 08/17/2006, and renamed POLICY 103‐1 in 2016, State Department of Education Content and Performance Standards for physical education.

Physical Activity Policy: Beginning in 2010–2011, students in all grades are required to have at least 20 minutes a day of supervised recess.

Applicable statutes and sources: State Department of Education Content and Performance Standards for physical education.

Idaho

Physical Education Policy: Effective in 2006, elementary and middle schools are required to offer physical education.

Applicable statutes and sources: IDAPA 08.02.03.104, State Board of Education Administrative Rules 08.02.02.104.01‐04.

Physical Activity Policy: Beginning in 1963, schools are required to offer physical fitness.

Applicable statutes and sources: Idaho Code § 33–1605.

Illinois

Physical Education Policy: Since 1996, per 105 ILCS 5/27‐5 and 23 Ill. Adm. Code 1.420, all students are required to engage in daily physical education.

Applicable statutes and sources: 105 ILCS 5/27‐6; 105 ILCS 5/27‐5; 23 Ill. Adm. Code 1.420.

Indiana

Physical Education Policy: Beginning in 2010, per 511 IAC 6.1‐5‐2.6, the elementary school curriculum is required to include physical education. Beginning in 2010, per 511 IAC 6.1‐5‐3.6, the middle school curriculum is required to include physical education. The Indiana State Board of Education recommends that students in grades 1–3 participate in 105 minutes per week of motor skills development and health/safety education, students in grades 4–6 participate in 75 minutes of PE per week, and students in grades 6–8, when grade 6 is included in a middle school, participate in 100 minutes of PE per week.

Applicable statutes and sources: 511 IAC 6.1‐5‐2.5; 511 IAC 6–7.1‐5; 511 IAC 6.1‐5‐3.6, Indiana State Board of Education.

Physical Activity Policy: Beginning in 2005–2006, per Burns Ind. Code Ann. § 20‐30‐5‐7, each school shall include physical fitness as part of the curriculum. Beginning in 2006–2007, per Burns Ind. Code Ann. § 20‐30‐5‐7.5, elementary schools are required to provide daily physical activity, which may include recess.

Applicable statutes and sources: Burns Ind. Code Ann. § 20‐30‐5‐7, Burns Ind. Code Ann. § 20‐30‐5‐7.5.

Iowa

Physical Education Policy: Beginning in 1986, per Iowa Code § 256.11, physical education is required to be taught in grades 1 – 8.

Applicable statute or sources: Iowa Code § 256.11.

Physical Activity Policy: Beginning in 2009, per Iowa Code § 256.11, students in grades K‐5 are required to engage in 150 minutes per week (30 minutes per day) of physical activity and students in grades 6–12 are required to engage in 120 minutes per week of physical activity.

Applicable statutes and sources: Iowa Code § 256.11.

Kansas

Physical Education Policy: Elementary and secondary schools, to receive accreditation, are required to offer physical education.

Applicable statutes and sources: K.A.R. § 91‐31‐32; K.A.R. § 91‐31‐35. The available state statutes date back only to 2005. According to the Shape of the Nation and Trust for America's Health reports, the requirement dates back to at least 1993.

Kentucky

Physical Education Policy: Beginning in 2006–2007, students in grades K‐8 are required to participate in PE as part of the state curriculum.

Applicable statute and sources: 704 KAR 3:303; 704 KAR 3:305.

Louisiana

Physical Education Policy: Since 1980, per La. R.S. § 17:276, LA secondary schools have been required to offer physical education courses.

Applicable statutes and sources: La. R.S. § 17:276.

Physical Activity Policy: Beginning in 2004–2005, per 2004 La. ALS 734, schools that include grades K‐6 are required to provide 150 minutes of moderate‐to‐vigorous physical activity for students. In 2009, per La. R.S. § 17:17.1, schools that include grades K‐8 are required to provide 150 minutes of moderate‐to‐vigorous physical activity for students.

Applicable statutes and sources: 2004 La. ALS 734, 2009 La. ALS 286, La. R.S. § 17:17.1, State Bulletins 102 and 103.

Maine

Physical Education Policy: Since 1983, elementary and secondary schools shall offer physical education. Since 2009, elementary schools must offer physical education as part of the course of study.

Effective January 1985, per CMR 94‐348‐004,physical education was required in secondary schools. Effective 2007, per CMR 05‐071‐132 and 20‐A M.R.S.A. § 1454, requires physical education instruction for prek – 8. Since 2009, per 2009 ch. 313, § 11 (AMD) and 20‐A M.R.S. § 4711,elementary, middle and junior high (per 20‐A M.R.S. § 4712) and secondary schools (per 20‐A M.R.S. § 4721)schools must provide for the instruction of all students….physical education.

Applicable statutes and sources: 20‐A M.R.S. § 4711; 20‐A M.R.S. § 4723; 2001 Me. ALS 454.

Maryland

Physical Education Policy: Since 1996, per Md. EDUCATION Code Ann. § 7–409, each school shall provide physical education to all students in grades K – 12.

Applicable statutes and sources: Md. EDUCATION Code Ann. § 7–409.

Massachusetts

Physical Education Policy: Since 1962, per ALM GL ch. 71, § 1, physical education shall be taught as a required subject in all grades for all students in the public.

Applicable statutes and sources: ALM GL ch. 71, § 1; ALM GL ch. 71, § 3.

Michigan

Physical Education Policy: Since 1976, per MCLS § 380.1502, physical education shall be provided to all students in all public schools. Beginning in 2003–2004, the state Board of Education recommends that all schools offer PE for 150 minutes per week for grades K‐5 and for 225 minutes per week for grades 6–8.

Applicable statutes and sources: MCLS § 380.1502; MCLS § 380.1278a.

Minnesota

Physical Education Policy: Beginning in 2005–2006 per 2005 Minn. ALS 5, Statute 120A‐22, students in grades 1–10 are required to participate in physical education. Beginning in the 2016–2017 school year, per Minn. Stat. § 120B.021, the department must adopt the most recent National Association of Sport and Physical Education standards and benchmarks for grades K‐12.

Applicable statutes and sources: 2005 Minn. ALS 5, Statute 120A‐22; Minn. Stat. § 120B.021.

Mississippi

Physical Education Policy: Between 2002–2003 and 2006–2007, the legislature recommends that students in grades K‐8 participate in PE for 150 minutes per week. Beginning in 2007–2008, per Miss. Code Ann. § 37‐13‐134, students in grades K‐8 are required to participate in physical activity instruction for 150 minutes per week.

Applicable statutes: Miss. Code Ann. § 37‐13‐134; 2002 Miss. ALS 585; Miss. Code Ann. § 37‐11‐8.

Missouri

Physical Education Policy: Effective 1995, per § 161.102 R.S.Mo, schools are required to provide PE to all students in all public schools. Beginning in 2001–2002, per 5 CSR 20–100.255, students in grades K‐5 are required to participate in 50 minutes per week of PE and students in grades 6–8 are required to receive 3000 minutes of instruction in PE per year. Beginning with the school year 2010–11, per § 167.720 R.S.Mo., school districts shall ensure that students in elementary schools participate in moderate physical activity for the entire school year of an average of 150 minutes per five‐day school week, or an average of 30 minutes per day. Students in middle schools may, at the school's discretion, participate in at least 225 minutes of physical activity per school week.

Applicable statutes: § 161.102 R.S.Mo. and Missouri School Improvement Program; § 167.720 R.S.Mo; 5 CSR 20–100.255; 200 MSGR 126.

Montana

Physical Education Policy: Since 1998, schools are required to offer PE.

Applicable statutes and sources: MONT. ADMIN. R. 10.54.2501.

Nebraska

Physical Education Policy: Per Nebraska Admin. Code Title 92, Ch. 10 004.03A9, schools are required to offer physical education in elementary and middle schools.

Applicable statutes and sources: Nebraska Admin. Code Title 92, Ch. 10‐004.03A9; Nebraska Admin. Code Title 92, Ch. 10 004.02A6.

Nevada

Physical Education Policy: Beginning in 1999–2000, PE “must be taught as applicable for grade levels and to the extent practicable in all public schools.”

Applicable statutes and sources: Nev. Rev. Stat. Ann. § 389.018; CHAPTER: 389 LCB File No. R037‐12A.

New Hampshire

Physical Education Policy: Beginning in 1975 or possibly earlier, per RSA 189:10, PE is required to be taught to students in all grades. Beginning in 2004–2005, per RSA 189:11‐a, the school boards of each school district were required to adopt a policy recommending daily PA. Effective March 2014, per 231 NH GOV REG 24, NH readopted Ed 306.41 to require physical education in k‐ 12.

Applicable statutes and sources: RSA 189:10; RSA 189:11‐a; 231 NH GOV REG 24.

New Jersey

Physical Education Policy: 1967, per N.J. Stat. § 18A:35‐5 and N.J. Stat. § 18A:35‐8, students are required to take courses in PE and health in grades 1–12 for a combination of 150 minutes per week. Local school districts determine the percentage of the 150 minutes that is allocated to PE.

Applicable statutes and sources: N.J. Stat. § 18A:35‐7; N.J. Stat. § 18A:35‐8; N.J. Stat. § 18A:35‐5.

New Mexico

Physical Education Policy: Beginning in 2005–2006, schools are required to provide PE in grades 1–8. Prior to 2005, schools are required to provide physical fitness to grades 4–8.

Applicable statutes and sources: N.M. Stat. Ann. § 22‐13‐1.

New York

Physical Education Policy: Effective 1983, per 8 NYCRR § 135.4, students in grades K‐3 must participate in PE daily and students in grades 4–6 must participate in PE at least 3 times per week. Students in grades K‐6 must participate in at least 120 minutes per week. Schools are required to offer PE to students in grades 7–8 at least 3 times per week in one semester and 2 times per week in the other semester.

Applicable statutes and sources: 8 NYCRR § 135.4.

North

Carolina

Physical Education Policy: Since 1995 per N.C. Gen. Stat. § 115C‐12, PE shall be offered to every student in the state. By the 2006 school year Policy ID Number:GCS‐S‐000 shall be fully implemented. Per Policy ID Number:GCS‐S‐000, the recommended minimum amount of PE is 150 minutes per week for students in grades K‐5 and 225 minutes per week for students in grades 6–8.

Applicable statutes and sources: Policy ID Number:GCS‐S‐000, N.C. Gen. Stat. § 115C‐12.

North

Dakota

Physical Education Policy: Since 2001, public and nonpublic elementary and middle schools (grades K – 8) shall provide students instruction in physical education.

Applicable statutes and sources: N.D. Cent. Code, § 15.1‐21‐01.

Ohio

Physical Education Policy: Since 1993, schools are required to offer PE for students in all grades.

Applicable statutes and sources: OAC Ann. 3301‐35‐04; ORC Ann. 3313.60, 1991 Ohio SB 195.

Oklahoma

Physical Education Policy: Beginning in 2006–2007, schools are required to provide PE for elementary school students for at least 60 minutes per week.

Applicable statutes and sources: 70 Okl. St. § 11–103.9.

Oregon

Physical Education Policy: Beginning in 1997–1998, per Or. Admin. R. 581‐022‐1210, schools are required to offer PE to students in all grades. Effective the 2017–2018 school year, per ORS § 329.496 every public school student in kindergarten through grade 8 shall participate in physical education for the entire school year. Students in kindergarten through grade 5 shall participate in physical education for at least 150 minutes during each school week. Students in grades 6 through 8 shall participate in physical education for at least 225 minutes during each school week.

Applicable statutes and sources: Or. Admin. R. 581‐022‐1210; ORS 329.045; ORS § 329.496.

Pennsylvania

Physical Education Policy: PE is required to be taught to students in all grades.

Applicable statutes and sources: 22 Pa. Code § 4.27; 022 PA Code 4.21

Rhode Island

Physical Education Policy: Students are required to participate in PE for at least 20 minutes each day in grades 1–12. Recess cannot count towards the PE requirements.

Applicable statutes and sources: Rules and Regulations of School Health Programs; R.I. Gen. Laws § 16‐22‐4

South

Carolina

Physical Education Policy: Beginning in 2006–07, per S.C. Code Ann. § 59‐10‐10, students in grades K‐5 must be provided a minimum of 150 minutes a week of physical education and physical activity. In 2006–07, a minimum of 60 minutes a week must be provided in physical education, and as Section 59‐10‐20 is phased in, the minimum time for physical education must be increased to 90 minutes a week.

Applicable statutes and sources: S.C. Code Ann. § 59‐10‐10; S.C. Code Ann. § 59‐10‐10.

South

Dakota

Physical Education Policy: Schools are not required to provide PE to students in grades K‐8.

Applicable statutes and sources: N/A

Tennessee

Physical Education Policy: Per Tenn. Comp. R. & Regs. R. 0520‐01‐03‐.05., schools are required to offer PE to students in grades K‐8. Beginning in 2006–2007, at least 90 minutes of physical activity must be provided throughout the day for students in grades K‐8.

Applicable statutes and sources: Tenn. Comp. R. & Regs. R. 0520‐01‐03‐.05; T.C.A. § 49‐6‐1021.

Texas

Physical Education Policy: Schools must offer PE to students in grades K‐5 since at least 1993 per 19 TAC § 74.2. Beginning in 2002–2003, per Senate Bill 19 passed in 2001, students in grades K‐6 must participate in at least 30 minutes per day or 135 minutes per week of physical activity (PA). In 2004–2005, schools are required to offer PE to students in grades 6–8. Beginning in 2006–2007, per Senate Bill 42 passed in 2005, students in grades 6–8 are required to participate in PE for at least 30 minutes per day or 135 minutes per week or 225 minutes per fortnight. The requirements for grades K‐5 refer to physical activity and permit recess to contribute to this requirement. Since December 2009 per 19 TAC § 74.37, On a weekly basis, at least 50% of a physical education class shall be used for actual student physical activity and the activity shall be, to the extent practicable, at a moderate or vigorous level.

Applicable statutes and sources: Administrative Codes 74.2, 74.32, and 74.3, Tex. Educ. Code § 28.002; 19 TAC § 74.37.

Utah

Physical Education Policy: Students in grades K‐6 are required to receive instruction in PE.

Applicable statutes and sources: U.A.C. R277‐700‐4.

Vermont

Physical Education Policy: Schools are required to offer PE. The Nutrition and Physical Fitness Guidelines, established in 2005, recommend 150 minutes per week for elementary school students and 225 minutes per week for middle school students.

Applicable statutes and sources: 16 V.S.A. § 906.

Virginia

Physical Education Policy: Beginning in 1999–2000, PE must be taught as part of the curriculum in all grades. Beginning in 2008–2009, a program of physical fitness, which can include PE, must be available to all students for at least 150 minutes per week.

Applicable statutes and sources: Va. Code Ann. § 22.1‐253.13:1; Code 22.1‐207; Code 22.1‐200.

Washington

Physical Education Policy: Students in grades 1–8 shall receive instruction in PE. By 2010, per Rev. Code Wash. (ARCW) § 28A.210.365, all students in grades one through eight should have at least one hundred fifty minutes of quality physical education every week

Applicable statutes and sources: Rev. Code Wash. (ARCW) § 28A.230.040; Rev. Code Wash. (ARCW) § 28A.210.365.

West

Virginia

Physical Education Policy: Elementary and middle schools are required to provide PE. Beginning in 2005–2006, per § 18‐2‐7a., students in grades K‐5 are required to participate in at least 90 minutes of PE per week and students in grades 6–8 are required to participate in daily PE for one full period for one semester.

Applicable statutes and sources: W. Va. Code § 18‐2‐7a

Wisconsin

Physical Education Policy: Beginning in 1988–1989, students in grades K‐6 are required to participate in PE at least 3 days per week and students in middle school are required to participate in PE at least once per week.

Applicable statutes and sources: Wis. Stat. § 121.02, 1987 Wis. ALS 27.

Wyoming

Physical Education Policy: Beginning in 1998–1999, each school district shall provide educational programs that include PE. Students in grades 1–8 are required to be proficient in PE.

Applicable statutes and sources: Wyo. Stat. § 21‐9‐101

Open in a separate window

Note: Laws passed before 1998 are described if they were still in effect during the 1998–1999 school year. This table updates a table in Frisvold, Cawley, and Meyerhofer (2013).10

References

1. Council on Sports Medicine and Fitness; Council on School Health. Active healthy living: prevention of childhood obesity through increased physical activity. Pediatrics. 2006;117(5):1834‐1842. 10.1542/peds.2006-0472 [PubMed] [CrossRef] [Google Scholar]

2. Pate RR, Davis MG, Robinson TN, Stone EJ, McKenzie TL, Young JC. Promoting physical activity in children and youth: a leadership role for schools: a scientific statement from the American Heart Association Council on Nutrition, Physical Activity, and Metabolism (Physical Activity Committee) in collaboration with the Councils on Cardiovascular Disease in the Young and Cardiovascular Nursing. Circulation. 2006;114(11):1214‐1224. 10.1161/CIRCULATIONAHA.106.177052 [PubMed] [CrossRef] [Google Scholar]

3. Committee on Physical Activity and Physical Education in the School Environment; Food and Nutrition Board; Institute of Medicine. Educating the Student Body: Taking Physical Activity and Physical Education to School. Kohl HW III, Cook HD, eds. National Academies Press (US); 2013. [PubMed] [Google Scholar]

4. Blundell JE, Stubbs RJ, Hughes DA, Whybrow S, King NA. Cross talk between physical activity and appetite control: does physical activity stimulate appetite?Proc Nutr Soc. 2003;62(3):651‐661. 10.1079/PNS2003286 [PubMed] [CrossRef] [Google Scholar]

5. Speakman JR, Selman C. Physical activity and resting metabolic rate. Proc Nutr Soc. 2003;62(3):621‐634. 10.1079/PNS2003282 [PubMed] [CrossRef] [Google Scholar]

6. Casazza K, Fontaine KR, Astrup A, etal. Myths, presumptions, and facts about obesity. N Engl J Med. 2013;368(5):446‐454. 10.1056/NEJMsa1208051 [PMC free article] [PubMed] [CrossRef] [Google Scholar]

7. Brown T, Moore TH, Hooper L, etal. Interventions for preventing obesity in children. Cochrane Database Syst Rev. 2019;7(7):CD001871. 10.1002/14651858.CD001871.pub4 [PMC free article] [PubMed] [CrossRef] [Google Scholar]

8. American Academy of Pediatrics. AAP publications reaffirmed or retired. Pediatrics. 2016;138(2):e20161650. 10.1542/peds.2016-1650 [CrossRef] [Google Scholar]

9. Cawley J, Meyerhoefer C, Newhouse D. The impact of state physical education requirements on youth physical activity and overweight. Health Econ. 2007;16(12):1287‐1301. 10.1002/hec.1218 [PubMed] [CrossRef] [Google Scholar]

10. Cawley J, Frisvold D, Meyerhoefer C. The impact of physical education on obesity among elementary school children. J Health Econ. 2013;32(4):743‐755. 10.1016/j.jhealeco.2013.04.006 [PubMed] [CrossRef] [Google Scholar]

11. Bednar S, Rouse K. The effect of physical education on children's body weight and human capital: new evidence from the ECLS‐K:2011. Health Econ. 2020;29(4):393‐405. 10.1002/hec.3990 [PubMed] [CrossRef] [Google Scholar]

12. Lexis Library. LexisNexis;2021.

13. School health policy database. State Policy Database Developed by the National Association of State Boards of Education (NASBE). 2020. Accessed March 4, 2020. https://statepolicies.nasbe.org/health

14. Healthy Schools Act of 2010, DC Official Code § 38–821.01(2010). Accessed May 14, 2021. https://doee.dc.gov/publication/healthy‐schools‐act‐2010

15. Early childhood longitudinal (ECLS) program. National Center for Education Statistics. Accessed May 14, 2021. https://nces.ed.gov/ecls/kinderdatainformation.asp

16. Barlow SE; Committee Expert. Expert committee recommendations regarding the prevention, assessment, and treatment of child and adolescent overweight and obesity: summary report. Pediatrics. 2007;120(Suppl 4):S164‐S192. 10.1542/peds.2007-2329C [PubMed] [CrossRef] [Google Scholar]

17. Flegal KM, Wei R, Ogden CL, Freedman DS, Johnson CL, Curtin LR. Characterizing extreme values of body mass index–for‐age by using the 2000 Centers for Disease Control and Prevention growth charts. Am J Clin Nutr. 2009;90(5):1314‐1320. 10.3945/ajcn.2009.28335 [PubMed] [CrossRef] [Google Scholar]

18. Strumpf EC, Harper S, Kaufman JS. Fixed effects and difference‐in‐differences. In: Oakes JM, Kaufman JS, eds. Methods in Social Epidemiology. John Wiley & Sons; 2017. [Google Scholar]

19. Angrist JD, Pischke JS. Mostly Harmless Econometrics: An Empiricist's Companion. 1st ed.Princeton University Press; 2008. [Google Scholar]

20. Wong VC, Valentine JC, Miller‐Bains K. Empirical performance of covariates in education observational studies. J Res Educ Eff. 2017;10(1):207‐236. 10.1080/19345747.2016.1164781 [CrossRef] [Google Scholar]

21. Cameron AC, Miller DL. A practitioner's guide to cluster‐robust inference. J Hum Resour. 2015;50(2):317‐372. [Google Scholar]

22. Texas State Legislature. S 19, Legislative Session 77(R), Section28 (2001).https://capitol.texas.gov/tlodocs/77R/billtext/html/SB00019F.htm

23. Bartlett J. Interval regression with heteroskedastic errors. The Stats Geek. February 20, 2015. Accessed July 2, 2022. https://thestatsgeek.com/2015/02/20/interval‐regression‐with‐heteroskedastic‐errors/

24. Bowling A, Davison K, Haneuse S, Beardslee W, Miller DP. ADHD medication, dietary patterns, physical activity, and BMI in children: a longitudinal analysis of the ECLS‐K study. Obesity. 2017;25(10):1802‐1808. 10.1002/oby.21949 [PubMed] [CrossRef] [Google Scholar]

25. Jia P, Xue H, Cheng X, Wang Y, Wang Y. Association of neighborhood built environments with childhood obesity: evidence from a 9‐year longitudinal, nationally representative survey in the US. Environ Int. 2019;128:158‐164. 10.1016/j.envint.2019.03.067 [PubMed] [CrossRef] [Google Scholar]

26. Katzmarzyk PT, Denstel KD, Beals K, etal. Results from the United States 2018 Report Card on Physical Activity for Children and Youth. J Phys Act Health. 2018;15(s2):S422‐S424. 10.1123/jpah.2018-0476 [PubMed] [CrossRef] [Google Scholar]

27. World Health Organization. School Policy Framework: Implementation of the WHO Global Strategy on Diet, Physical Activity and Health. World Health Organization; 2008. [Google Scholar]

28. Thompson HR, Johnson RC, Madsen KA, Fuller B. Impact of physical education litigation on fifth graders’ cardio–respiratory fitness, California, 2007–2018. Am J Public Health. 2019;109(11):1557‐1563. 10.2105/AJPH.2019.305264 [PMC free article] [PubMed] [CrossRef] [Google Scholar]

29. Rothstein R, Jacobsen R, Wilder T. Grading Education: Getting Accountability Right. Teachers College Press & Economic Policy Institute; 2008. [Google Scholar]

30. Carlson SA, Fulton JE, Lee SM, etal. Physical education and academic achievement in elementary school: data from the early childhood longitudinal study. Am J Public Health. 2008;98(4):721‐727. 10.2105/AJPH.2007.117176 [PMC free article] [PubMed] [CrossRef] [Google Scholar]

31. Zach S, Shoval E, Lidor R. Physical education and academic achievement—literature review 1997–2015. J Curric Stud. 2017;49(5):703‐721. 10.1080/00220272.2016.1234649 [CrossRef] [Google Scholar]

32. Rasberry CN, Lee SM, Robin L, etal. The association between school‐based physical activity, including physical education, and academic performance: a systematic review of the literature. Prev Med. 2011;52:S10‐S20. 10.1016/j.ypmed.2011.01.027 [PubMed] [CrossRef] [Google Scholar]

33. Thompson HR, Singh BK, Reed A, etal. Impact of litigation on compliance with California physical education laws in elementary schools. J Phys Act Health. 2018;15(10):721‐729. 10.1123/jpah.2017-0307 [PubMed] [CrossRef] [Google Scholar]

34. Nader PR. Frequency and intensity of activity of third‐grade children in physical education. Arch Pediatr Adolesc Med. 2003;157(2):185‐190. [PubMed] [Google Scholar]

35. Hall KD, Gortmaker S, Lott M, Yang YC. From calories to weight change in children and adults: the state of the science. Healthy Eat Res. Published online June 2016. [Google Scholar]

36. Swinburn B, Sacks G, Ravussin E. Increased food energy supply is more than sufficient to explain the US epidemic of obesity. Am J Clin Nutr. 2009:90(6):1453‐1456 [PubMed] [Google Scholar]

37. Cutler DM, Glaeser EL, Shapiro JM. Why have Americans become more obese?J Econ Perspect. 2003;17:93‐118. [Google Scholar]

38. Miller WC, Koceja DM, Hamilton EJ. A meta‐analysis of the past 25 years of weight loss research using diet, exercise or diet plus exercise intervention. Int J Obes. 1997;21(10):941‐947. 10.1038/sj.ijo.0800499 [PubMed] [CrossRef] [Google Scholar]

39. Hook JV, Altman CE. Competitive food sales in schools and childhood obesity: a longitudinal study. Sociol Educ. 2012;85(1):23‐39. 10.1177/0038040711417011 [PMC free article] [PubMed] [CrossRef] [Google Scholar]

40. Kiesel K, Ji M. Did state‐mandated restrictions on sugar‐sweetened drinks in California high schools increase soda purchases in school neighborhoods?Appl Econ Perspect Policy. 2021;43(4):1443‐1475. 10.1002/aepp.13137 [CrossRef] [Google Scholar]

41. Gleason PM, Suitor C. Children's Diets in the Mid‐1990s: Dietary Intake and Its Relationships with School Meal Participation. US Department of Agriculture, Food and Nutrition Service; 2001. [Google Scholar]

42. von Hippel PT, Workman J. From kindergarten through second grade, U.S. children's obesity prevalence grows only during summer vacations. Obesity. 2016;24(11):2296‐2300. 10.1002/oby.21613 [PubMed] [CrossRef] [Google Scholar]

43. Brazendale K, Beets MW, Weaver RG, etal. Understanding differences between summer vs. school obesogenic behaviors of children: the structured days hypothesis. Int J Behav Nutr Phys Act. 2017;14(1):100. 10.1186/s12966-017-0555-2 [PMC free article] [PubMed] [CrossRef] [Google Scholar]

Articles from The Milbank Quarterly are provided here courtesy of Milbank Memorial Fund