Cross-sectional research

Relationship between physical activity, sport or fitness and cognitive testing or academic test results in children.

Terms

  • +: positive association
  • 0: no association
  • Ass: association
  • BMI: Body mass index
  • EAT: Eating Among Teens
  • CAT6: California Achievement Tests version 6
  • CST: California Standards Tests
  • GPA: grade point average
  • ISAT: Illinois Standards Achievement Test
  • K-BIT: Kaufman Brief Intelligence Test
  • LTAQ: Leisure Time Activity Questionnaire
  • MCAS: Massachusetts Comprehensive Assessment Systems
  • PA: physical activity
  • PE: physical education.

Bass et al., 2013, (Bass et al., 2013) Illinois State University, USA

Sample

Participants were students from a public middle school (grades 6–8) in central Illinois.

Methods

Correlational design.

Measures

  • The FITNESSGRAM test battery assessed students (n = 838) in the five components of health-related fitness. The Illinois Standardised Achievement Test (ISAT) was used to assess academic achievement in reading and math.

Results

The largest correlations were seen for aerobic fitness and muscular endurance (ranging from 0.12 to 0.27, all p < 0.05). Boys in the Healthy Fitness Zone (HFZ) for aerobic fitness or muscular endurance were 2.5–3 times more likely to pass their math or reading exams. Girls in the HFZ for aerobic fitness were approximately 2–4 times as likely to meet or exceed reading and math test standards.

Association

+

Study limitations

First, this was a cross- sectional study, and causation cannot be established.

Pirrie & Lodewyk, 2012, (Pirrie & Lodewyk, 2012) Brock University, Ontario, Canada

Sample

Participants were two classes of fourth-grade students in Ontario (n = 40).

Methods

The study recruited 40 fourth-grade students from two classes in two independent schools (one semi-urban [n = 19] and one rural [n = 21]) in a school district situated in south-western Ontario, Canada.

Measures

  • Students in the two classes completed standardised tests for each cognitive process both after no physical activity and following it (20-minute MVPA within a 45-minute lesson).

Results

The results indicate that performance on the planning test significantly improved after physical activity (p<0.001), controlling for sequence and habituation/retesting effects. No improvement was observed for attention, simultaneous processing, or successive processing.

Association

+ planning
0 attention
0 simultaneous processing
0 successive processing

Study limitations

For all of the cognitive tests, test administration timing post-activity may have impacted the strength of results.

Jonker, Elferink-Gemser, Toering, Lyons, & Visscher, 2010, (Jonker et al., 2010) University Medical Center Groningen

Sample

128 elite soccer players aged 12-16 years, and 164 aged-matched controls (typical students).

Methods

Elite youth aged matched with typical students.

Measures

  • Self-completed questionnaire.

Results

Findings suggests that the relatively stronger self-regulatory skills (self-conscious, goal-oriented, and problem-focused behaviours) reported by the elite youth soccer players may be essential for performance at the highest levels of sport competition and in academia.

Association

+

Study limitations

Caution is needed regarding this proposition, as it may also be the case that the elite youth soccer players are high achievers in sport and education because of an inherent ability to self-regulate. In other words, do the elite youth soccer players compete at a high level because their self-regulatory skills were developed through sport, or because these skills were inherent?

Jonker, Elferink-Gemser, & Visscher, 2009, (Jonker et al., 2009) University Medical Center Groningen

Sample

400 talented athletes participated in this study.

Methods

Athletes were classified as ‘talented athletes’ on the basis of their qualifications by the Netherlands Olympic Committee and Sports Federation (NOC*NSF) and were therefore all part of a talent program.

Measures

  • Self-completed questionnaire.

Results

When compared with the national average, the athletes in 2006-07 attended pre- university classes more often (χ2 = 57.001, p<.05). Of the 2006-07 athletes, a higher percentage participated in pre-university programs compared with that of athletes in 1992-93 (χ2 (1, n = 400) = 32.003, p<.05), whereas the national averages showed stability (χ2 = .325, p>.05).

Association

+

Study limitations

Within the Dutch educational system, all students are used to the possibility to ask for extra supervision or help by a mentor if required. Therefore, talented athletes who are also high achievers academically may be based on differences in instruction caused by the special provisions offered to them.

Kristjánsson, Sigfúsdóttir, Allegrante, & Helgason, 2009, (Kristjánsson et al., 2009) Reykjavik University, Iceland

Sample

5,810 adolescents.

Methods

Self-completed questionnaire.

Measures

  • Academic achievement
  • School contentment
  • Body mass Index
  • Sedentary lifestyle
  • Physical activity.

Results

Model explained 36% of the variance in academic achievement and 24% in school contentment. BMI and sedentary lifestyle were negatively related to school contentment and academic achievement, but physical activity was positively related to school contentment and academic achievement (P< .01). School contentment was strongly related to academic achievement but only a weak mediator of the health behaviour indicators.

Findings may inform the efforts to improve academic achievement and the general health status of youth.

Association

+

Study limitations

Some of the measured relationships are quite weak, particularly those stemming from BMI and sedentary lifestyle.

Kwak et al., 2009, Unit for Preventive Nutrition, Sweden (Kwak et al., 2009)

Sample

Swedish 9th-grade students (n = 232; mean age = 16 years; 52% girls).

Methods

Groups of pupils, within each school (n = 42), were randomly selected proportional to the sizes of the respective schools.

Measures

  • School grades, pubertal phase, skinfold thickness, cardiovascular fitness, and physical activity were measured objectively. Mother’s education, family structure, and parental monitoring were self-reported. Data were analysed with linear regression analyses.

Results

In girls, academic achievement was associated with vigorous physical activity and not mediated by fitness, whereas in boys only fitness was associated with academic achievement.

Association

+

Study limitations

Further studies are necessary to investigate the potential longitudinal effect of vigorous physical activity on academic achievement, the role of fitness herein and the implications of these findings for schools. The use of accelerometers, even though seen as a ‘‘golden standard’’; they are limited in capturing any activities with little displacement of the body, such as cycling and snow-boarding.

LeBlanc et al., 2012, (LeBlanc et al., 2012) Southeastern Louisiana University, USA

Sample

Participants were 1,963 children in fourth to sixth grades.

Methods

Correlational design. Adiposity was assessed by calculating body mass index (BMI) percentile and percent body fat and academic achievement with statewide standardised tests in four content areas. Socioeconomic status and age were control variables.

Measures

  • Children wore an accelerometer for three days to provide objective measurement of physical activity. In addition, the association between weight status and academic achievement was examined by comparing children who could be classified as “extremely obese” and the rest of the sample, as well as comparing children who could be classified as normal weight, overweight, or obese. Extreme obesity was defined as >1.2 times the 95th percentile

Results

These results do not support the hypotheses that increased adiposity is associated with decreased academic achievement or that greater physical activity is related to improved achievement.

Association

0

Study limitations

These results are limited by methodological weaknesses, especially the use of cross-sectional data.

Morales, Pellicer-Chenoll, Garcia-Masso, Gomis, & Gonzalez, 2011,  Universitat Ramon Llull, Barcelona, Spain (Morales et al., 2011)

Sample

284 students (158 girls, 126 boys) with an average age of 14.7 years participated.

Methods

Self-completed survey instrument

Measures

  • Physical activity measures
  • Student academic records

Results

Results showed that there was a linear relationship between academic performance and physical activity; nevertheless, there was a trend to stronger correlation when modelling the relationship between these variables with a quadratic equation.

Association

+

Study limitations

Did not directly account for whether academic performance and physical activity might be better explained with a second-order equation.

Fox, Barr-Anderson, et. al. 2010,(Fox et al.) University of Minnesota

Sample

31 middle and high schools in metropolitan Minnesota, n = 7746 children.

Methods

Students completed the EAT survey, demographic information, sport team participation and GPA questions. 

Measures

  • Sports team participation (on how many sports teams did you play in the last 12 months) and academic grades (GPA) (two grades achieved most often)
  • Self report of PA measured using LTEQ.

Results

High school girls: PA and sport team participation independently associated with higher GPA; high school boys sports team participation independently associated with higher GPA; middle school students PA and sports team participation combined association with higher GPA.

Association

+

Study limitations

All data were self report. 

Roberts, Freed, McCarthy, 2010, (Roberts, Freed, & McCarthy, 2010) University of California 

Sample

1989 children in Years 5, 7 and 9 attending middle- to high-income South Carolina school district public schools. 

Methods

Aerobic fitness, body weight, student demographic data, standardised test score data and school district demographic data were taken from school and district information. Parents reported additional demographic data.

Measures

  • Fitnessgram
  • demographic
  • overweight risk status (from CDC weight status cut-points)
  • California Achievement Tests version 6 (CAT6) and California Standards Tests (CST).
     

Results

Aerobic fitness significantly related to standardised test scores. BMI significantly inversely related to standardised test scores.

Association

+

Study limitations

Limitations to Fitnessgram as measure of aerobic fitness. Children’s efforts may have impacted upon Fitnessgram results. 

Carlson, Fulton et.al. 2008, (Carlson et al., 2008) Centers for Disease Control and Prevention

Sample

5,316 kindergarten children nationally representative  sample from longitudinal study.

Methods

Teachers reported PE. Children were given maths and reading tests. Demographics collected from parents via telephone.

Measures

  • PE minutes per week collected from teachers
  • Maths and reading scores on item response theory scale.

Results

Girls who were enrolled in higher amounts of PE achieved higher maths and reading scores.

Association

+ (girls)
0 (boys) 

Study limitations

Time spent in PE self report and no reliability or validity assessment of this measure.

Castelli, Hillman, et. al. 2007, (Castelli, Hillman, Buck, & Erwin, 2007)  University of Illinois

Sample

259 3rd and 5th grade children at four public schools.

Methods

Children completed fitness testing and ISAT at school. 

Measures

  • Fitnessgram (muscle fitness, aerobic capacity, body composition) during PE
  • ISAT.

Results

Physical fitness positively associated with academic achievement. BMI inversely related to academic achievement. Associations noted for total academic and maths and reading achievement. 

Association

+

Study limitations

Methods used for measuring fitness have limitations. Sampling not random.

Tremarche, Robinson, 2007, (Tremarche, Robinson, & Graham, 2007) Bridgewater State College

Sample

Convenience sample of 311 4th grade students attending two Massachusetts schools. 

Methods

Comparison of test results at two schools: School 1 providing 28 hours and School 2 providing 56 hours of PE per year.  

Measures

  • MCAS (maths and English language and arts)
  • School demographics.

Results

Average English and language arts score higher at school with PE time greater than school with lower PE time. 

No difference in maths score averages between scores. 

Association

+ (English and language arts)
0 (maths) 

Study limitations

Convenience sample of children tested. 

While many school demographics measured, other school characteristics may have influences results. 

Dollman, Boshhoff et al 2006, (Dollman J et al., 2006) University of South Australia

Sample

117 South Australian Primary Schools.

Methods

Principal (or representative) completed questionnaire. Academic attainment data received from the education department. School averages for numeracy and literacy calculated. 

Measures

  • Minutes each class spent in PE during previous week.

Results

Schools with high levels of time spent in PE do not have lower academic achievement despite spending less time in academic subjects. No difference in academic scores in relation to time spent in PE. 

Association

+ (improved learning per unit of time)

Study limitations

Low response rate of schools invited to participate in study (30%). Schools committed to PE may be more likely to participate in study. Did not account for quality of PE. School level data used.

Sigfusdottir, Kristjanson et. al. 2006, (Sigfusdottir et al., 2006) Reykjavik University

Sample

All secondary schools in Iceland sent questionnaires for children aged 14 and 15 (9th and 10th grade). 6,346 students in total. 

Methods

Data obtained from 2000 Icelandic study, ‘Youth in Iceland’. Self-completed survey instrument.  

Measures

  • Self report of academic achievement
  • Self report of height, weight and PA levels.

Results

PA was a significant predictor of academic achievement when controlling for other variables. Body mass index, diet and PA explained up to 24% of the variance in academic achievement when controlling for gender, parental education, family structure and absenteeism. 

Association

+

Study limitations

Height and weight self report. Self report of PA levels. Data of individuals who did not enter a height or weight were not included possibly biasing results. Self report of average grades may not have reflected actual grades.

Hillman, Castelli  et. al. 2005, (Hillman, Castelli, & Buck, 2005) University of Illinois

Sample

51 children and adults. 24 children recruited from Champaign elementary school system.

Methods

Fitness tested using Fitnessgram. K-Bit Cognitive task and EEG administered. Matching of high and low fit participants to assist controlling for demographics.

Measures

  • Demographics 
  • Fitnessgram
  • EEG using 10-120 system
  • Cognitive task (visual oddball paradigm)
  • K-BIT to measure IQ.

Results

High-fit children had significantly faster reaction times  than low-fit children to target stimuli. 

Association

+

Study limitations

Other factors not measured could account for differences. Small sample size. Field test of fitness rather than more accurate objective measure.

Lidner, 2002, (Lidner KJ, 2002) The University of Hong Kong

Sample

Two randomly selected classes from randomly selected high schools in Hong Kong. 1,447 students aged 13-17 years.

Methods

Self-completed survey instrument.

Measures

  • Academic records collected from schools
  • Self report questionnaire.

Results

Significant positive link between academic performance and PA participation. Significant  positive relationship between PA participation and by band level of students (school grouping based on primary academic achievement).

Association

+

Study limitations

No objective measure of PA used. 

Dwyer, Sallis et al  2001, (Dwyer T et al., 2001) University of Tasmania

Sample

Randomly selected, nationally representative sample of 7,961 Australian Schoolchildren aged 7-15 years.

Methods

Data collected by 10 data collectors in each Australian state as part of the Australian Schools Health and Fitness Survey in 1985. Ratings of scholastic ability were given for each participant by school representative. 

Measures

  • Field tests of PA and fitness measures collected by trained data collectors
  • School ratings of scholastic ability 
  • Questionnaire: self perceived academic ability, involvement in exercise and sport.
     

Results

School ratings of scholastic ability were significantly associated with physical fitness, capacity and activity. There were also weak but consistent associations between scholastic ability and field tests of muscular force, endurance and power. Non-consistent results of cardio-respiratory endurance. 

Association

+

Study limitations

Disparity between two cardio-respiratory endurance results may be due to possible measurement bias or confounding. Field tests may have been influenced by motivation of students to perform.

Field, Diego et al  2001, (Field T et al., 2001)  University of Miami School of Medicine

Sample

89 high school students.

Methods

Self-completed questionnaire which included behavioural and exercise measures.

Measures

  • Exercise regularity per week
  • Sports involvement
  • Grade point average.

Results

Students reporting a high level of exercise spent significantly more time in sport and higher grade point averages. 

Association

+

Study limitations

All measures were self report. Small number of study participants. 

Tremblay, Inman and Willms, 2000, (Tremblay, Inman, & Willms, 2000) University of New Brunswick  

Sample

74.3% of total population of grade 6 students in New Brunswick Canada (n=6856).

Methods

Data from the Elementary School Climate Study used. Children completed of and study questionnaire and this was linked to standardised achievement test data collected by the education department. 

Measures

  • Maths and reading scores
  • BMI self report
  • SES
  • Study questionnaire, four questions on PA participation.

Results

No significant relationship between PA and maths and reading scores. 

Association

0

Study limitations

BMI and PA data self-report. 

Dexter 1999, (Dexter T, 1999) University of Cambridge 

Sample

517 candidates from sample of 17 schools taking the General Certificate of Secondary Education (GCSE).

Methods

Review of records.

Measures

  • Academic ability calculated from Maths and English GCSE scores
  • GCSE PE score.

Results

Significant positive correlation between academic ability and sport performance.

Association

+

Study limitations

Sport performance measurement taken under test conditions may not reflect normal performance. 

Lidner 1999, (Lidner KJ, 1999) The University of Hong Kong

Sample

One or two randomly selected classes from randomly selected primary and high schools in Hong Kong. 4,690 children grades 5-12.

Methods

Age adapted self-completed survey instrument.

Measures

  • Sport participation survey instrument
  • Desired sport activities
  • Self perceived rating of academic performance
  • Self-perceived rating of sport and PA ability.

Results

Frequency and extent of sports participation significantly higher for student with high self-ratings of academic performance.

Association

+

Study limitations

Self-reported rating of academic performance used. Use of grouping students of primary school students based on academic scores to their secondary school may have affected their self-perceived academic success. 

Sillijer and Quirk 1997, (Silliker SA & Quirk JT, 1997) St Bonaventure University

Sample

123 high school students from five similar schools.

Methods

Counsellor identified students involved in soccer. Data collected on a data sheet by school counsellor in- season and out- of-season.

Measures

  • Grade point averages (GPA) for in and out of season.

Results

Participants had significantly higher GPA in-season than out-of-season. 

Association

+

Study limitations

Data collected only for soccer players. Small sample size. Schools not randomly selected. GPA may have been influenced by another seasonal factor. 

Dwyer, Blizzard et al 1996, (Dwyer T et al., 1996) The University of Tasmania

Sample

2,400 Australian randomly selected children from 9000 school children recruited into the ASHRS study from 109 schools.

Methods

Self administered questionnaire and field testing by trained personnel.

Measures

  • Skinfold thickness
  • Endurance fitness
  • Leisure activity 
  • Academic performance.
     

Results

PA and physical capacity were significantly positively related to scholastic rating. These associations remained after adjusting for relevant confounders.

Association

+

Study limitations

Motivation may have effected field testing results. 

Fisher, Juszczak et al  1996, (Fisher et al., 1996) North Shore University Hospital and Cornwell University Medical College

Sample

838 students in one school.

Methods

Self-completed questionnaires during gym class.

Measures

  • Sports questionnaire including number and type of sports and time spent in sport
  • Self report average grade.

Results

Time spent playing sport was not significantly associated with academic performance.

Association

0

Study limitations

All students were involved in at least one sport. Small sample. All measures self-report. Questionnaires distributed during gym class. Reliability and validity testing of sports questionnaire not reported. 

Pate, Heath et. Al 1996, (Pate, Heath, Dowda, & Trost, 1996) University of South Carolina

Sample

11,631 high school students.

Methods

Self-completed questionnaire.

Measures

  • Self perceived academic performance 
  • Level of exercise in last 2 weeks
  • Involvement in sports teams (community and school-based).

Results

High PA levels were significantly associated with participation in high levels of sport. Low activity was associated with low perception of academic performance.

Association

+

Study limitations

Measures were all self report. Perception of academic performance may not reflect actual academic performance.