Applied Research in Quality of Life

, Volume 6, Issue 3, pp 277–289

Physical Activity, Life Satisfaction, and Self-Rated Health of Middle School Students


    • Department of Community Medicine, School of MedicineWest Virginia University
  • Rebecca J. White
    • Philadelphia College of Osteopathic Medicine

DOI: 10.1007/s11482-010-9129-z

Cite this article as:
Zullig, K.J. & White, R.J. Applied Research Quality Life (2011) 6: 277. doi:10.1007/s11482-010-9129-z


Little research has examined the association between life satisfaction, self-rated health (SRH), and physical activity concurrently for middle school students. A convenience sample of 245 students in grades 7 and 8 was surveyed about physical activity, life satisfaction, and SRH using the U.S. Centers for Disease Control and Prevention (CDC) 2005 Middle School Youth Risk Behavior Survey. ANOVA analyses revealed significantly reduced life satisfaction for females who reported not engaging in vigorous physical activity during the past 7 days [p < .01, effect size (ES) = .75]. Significantly reduced life satisfaction was detected for both males (p < .001, ES = .66) and females (p < .0001, ES = .80) who reported not playing on sports teams. Additionally, logistic regression analyses showed the odds of reporting fair/poor SRH increased 5.4 times for males (CI = 1.30–22.39, p < .05) and 30.9 times for females (CI = 3.74–255.43, p < .001) who reported not playing on sports teams. Preliminary findings suggest physical activity and sports participation is associated with improved life satisfaction and SRH for middle school students. In addition, although some gender differences were observed, consistent findings for sports participation suggest sports participation may carry multiple social, mental, and physical benefits for youth.


Life satisfactionSelf-rated healthPhysical activitySports participationYouthAdolescents


Measures of life satisfaction are related, but distinct from physical and mental health measures (Diener et al. 1999; Huebner 1997; Zullig et al. 2005) and consistent with positive mental health orientations that define psychological well-being as more than an absence of psychopathology (Seligman and Csikszentmihalyi 2000). Physical activity has gained attention for its contributions to the quality of life, as well as its role in the prevention of premature morbidity.

For example, physical activity has well-documented physical, social, and psychological benefits for adolescents, including increased cardiorespiratory fitness (Aires et al. 2010), weight gain prevention (Simon et al. 2008), and decreased body fat and blood pressure (Farpour-Lambert et al. 2009). Long-term physical benefits of physical activity include improved fitness and bone health in adulthood (Kemper et al. 2000), as well as decreased risks of chronic diseases including diabetes (Helmrich et al. 1991) heart disease, and cancer (Booth et al. 2000). In addition, increased adolescent physical activity has also been linked to reduced depressive symptoms (Sanders et al. 2000), improvements in school performance (Field et al. 2001; Sibley and Etnier 2003), and improved self-image, self-esteem, and relationships (Field et al. 2001; Kirkcaldy et al. 2002). Conversely, adolescent sedentary behavior and low physical fitness have been shown to contribute not only to poor current health among youth but also to poor adulthood health outcomes, including chronic diseases, obesity, and increased cholesterol, among others (Hancox et al. 2004; Boreham et al. 2002).

Middle school children have been identified as an understudied population in adolescent physical activity research (McKenzie and Kahan 2004). This study furthers the extant literature base in physical activity by incorporating important outcomes such as life satisfaction and self-rated health (SRH) among middle school students. Irwin and colleagues (2002) suggest that the relationship between self-rated health status and subsequent health problems among adolescents has been explored infrequently because adolescents are generally regarded as healthy. However, there is developing evidence that similar relationships can be observed for adolescent (SRH). For instance, SRH in adolescence (Vingilis et al. 1998, 2002; Wade et al. 2000) is associated with personal, socio-environmental, behavioral, and psychological factors (e.g., health problems, disability, age, female status, income, smoking, and higher body mass index). In addition, reduced adolescent SRH has been associated with reduced life satisfaction (Zullig et al. 2005) and with increased alcohol and substance use (Zullig et al. 2004). In sum, there is developing evidence to suggest a relatively robust relationship between reduced SRH and a variety of health-compromising conditions among adolescents.

Although the majority of adolescent life satisfaction and SRH research has been conducted with older students, research is emerging in younger populations (e.g., Esch and Zullig 2008; Huebner et al. 2005; Kwan 2010). However, few studies were located that examined adolescent life satisfaction and physical activity. The few studies conducted with older, secondary students suggest decreased physical activity is associated with reduced life satisfaction and SRH (Piko and Keresztes 2006; Piko 2000; Poulsen et al. 2006, 2007; Thorlindsson et al. 1990; Valois et al. 2004). In the two most related studies, Valois et al. (2004) found reduced life satisfaction among high school males who did not engage in vigorous exercise during the past 7 days and among white females and males who did not play on a sports team run by their school, but did not assess SRH. Piko and Keresztes (2006) furthered this research by examining life satisfaction, SRH, and life goals among students 14–21 years of age as related to physical activity. These researchers note that although higher levels of physical activity were associated with increased life satisfaction and SRH, higher physical activity levels were related to extrinsic values (e.g., material success, fame, and image experiences), but not to intrinsic values (e.g., aspirations like self-acceptance, affiliation, health, or community feeling) in their sample. The speculation is that sport activity may thus serve as an important avenue of personal growth and subsequently drive healthier behavior, but sports participation was not directly measured by Piko and Keresztes (2006).

Thus, the purpose of this study was to explore the relationship between physical activity (including sports participation), life satisfaction, and self-rated health concurrently in a sample of middle school students. It was hypothesized that decreases in reported physical activity and sports participation would be related to reduced satisfaction with life and SRH.



During the spring of 2005, a convenience sample of 245 middle school students in grades 7 and 8 were selected from two public school districts to participate as part of a larger study investigating the test-retest reliability of the 2005 Middle School YRBS (MSYRBS) from the U.S. CDC (Zullig et al. 2006). Matched by key demographic data, schools were selected to participate in the study based on previous work in comprehensive health education curriculum deliberation and program evaluation (see Ubbes and Zullig 2008 for a review). Each school was given $250 to assist with student recruitment. Classroom-level sampling was done with second period classes to maximize student eligibility. Table 1 presents the sample demographics of the students in the study. The sample contains 136 females and 109 males. Most students reported being in 7th grade and the majority of the students were white. The demographics of this sample were similar to national distributions in gender and age, but not for race or ethnicity or grade per the 2000 U.S. Census Bureau. Specifically, 7th grade students reporting “white” race ethnicity were overrepresented.
Table 1

Sample demographics of students in grades 7–8 in the sample and national distribution


Sample distribution (%)

National distribution (%)
































Data Collection Procedures

All data collection procedures were approved by the referent university’s Human Subjects Review Board. Following methods from previous test-retest reliability studies (Brener et al. 1995, 2002; Popham 1993), two questionnaire scantrons were coded with the same unique number ranging from 1 to 250. Each pair of identically numbered scantrons were then placed into a single large envelope along with the MSYRBS instrument and distributed to each student participant during Time 1. Each student then removed and used one scantron. The envelope containing only the second identically numbered scantron, was then sealed and signed by the student across the seal. When survey administrators returned for Time 2 (14 days later), students received their signed and sealed envelope, removed the second scantron, and destroyed the envelope. Parent-notification forms were distributed at least 7 days in advance of survey administration. Parents who wanted their children to participate were required to sign and return the form, designating active consent to participate. The survey was administered in each school’s auditorium or cafeteria by trained data collectors, who emphasized anonymity, privacy, and confidentiality and took most students 30 minutes or less to complete. During Time 1, data collectors reminded students that they would be returning to the school in 2 weeks to ask students to complete a similar questionnaire.


The MSYRBS is a self-reported instrument. For this study, four questions composed the independent variables of physical activity resulting in substantial reliability [Kappas 65.6–75.9] (Zullig et al. 2006) for middle school students and are exactly the same as the items used on the U.S. national high school YRBS (Eaton et al. 2006).

Table 2 contains the study independent variables. Data from the nationally implemented YRBS reports a similar frequency of youth in grades 9–12 engaged in the same surveyed physical activity variables as the seventh and eighth grade students in this survey (CDC 2005). For example, approximately 84% of high school students from the national YRBS reported engaging in vigorous exercise 1+ days in the past 7 days, 76.5% reported watching 1+ hours of television on the average school day, 54.2% reported attending physical education class 1+ days during the average school week, and 56% reported participating on one or more sports teams (CDC 2005). Thus, middle school students in this sample reported engaging in slightly more vigorous physical activity (92%) and sports participation (67%) than high school students, which is consistent with age-related declines in adolescent physical activity from middle to high school in the U.S. (Caspersen et al. 2000). Data were consistent for the other two independent study variables.
Table 2

Physical activity variables and frequencies

Risk variable

Males n (%)

Females n (%)

Vigorous exercise (past 7 days)

0 days

8 (7.3)

12 (8.8)

1+ days

101 (92.7)

124 (91.2)

Television watching (average school day)

Less than 1 h

22 (20.2)

32 (23.5)

1+ hours

87 (79.8)

104 (76.5)

Physical education class attendance

0 days (average school week)

46 (42.2)

61 (44.9)

1+ days (average school week)

63 (57.8)

75 (55.1)

Sports team participation


73 (67.0)

92 (67.6)


36 (33.0)

44 (32.4)

Dependent Variables

Life Satisfaction

The first dependent variable in this study was measured using one question querying students about their overall life satisfaction from the Brief Multidimensional Students’ Life Satisfaction Scale (BMSLSS) and contained the following response options: ‘Terrible’, ‘Unhappy’, ‘Mostly dissatisfied’, ‘Mixed-about equally satisfied and dissatisfied’, ‘Mostly satisfied’, ‘Pleased’, and ‘Delighted’(Seligson et al. 2003, 2005). Although the other BMSLSS domains were not available for the present study, previous research has shown the overall item to correlate highly with the other five pooled BMSLSS domains (satisfaction with school, friends, living environment, family, and self) where between 49% and 64% of the variance in overall adolescent life satisfaction reports was explained by the other domains (Seligson et al. 2003, 2005). In addition, the BMSLSS scale alpha coefficient has been reported as .85 among middle school students (Seligson et al. 2003). Table 3 contains the frequency counts for the overall life satisfaction variable.
Table 3

Reported life satisfaction and self-rated health frequencies

Dependent variables

Males n (%)

Females n (%)

Life satisfaction


4 (3.7)

1 (0.7)


2 (1.8)

2 (1.5)

Mostly dissatisfied

5 (4.6)

3 (2.2)


16 (14.7)

34 (25.0)

Mostly satisfied

16 (14.7)

29 (21.3)


26 (23.8)

33 (24.3)


40 (36.7)

34 (25.0)

Self-Rated health


26 (23.9)

26 (19.1)

Very Good

44 (40.4)

54 (39.7)


29 (26.6)

45 (33.1)


7 (6.4)

9 (6.6)


3 (2.7)

2 (1.5)

Self-Rated Health

The second study dependent variable for this study was “In general, how would you describe your health?” with five response options: ‘Excellent’, ‘Very good’, ‘Good’, ‘Fair’, and ‘Poor’. For this study, SRH was dichotomized into Fair/Poor health and compared to the referent Excellent/Very Good/Good health rating and odds ratios (OR) are reported. The Kappa test-retest reliability coefficient for self-rated health was 61.8 in this sample and is detailed elsewhere (Zullig et al. 2006). Table 3 also contains the frequency counts for SRH.

Data Analysis

In this study, Time 1 survey administrations from each of the participating schools were combined for analysis. Of the 402 students selected to participate in this study, 249 (62%) completed the questionnaire during the first survey administration. Of these 249 students, 98% (n = 245) students provided complete information on all items.

All analyses were performed via PC-SAS. Multivariate analysis of variance (MANOVA) was initially conducted to determine if there was an overall significant difference among the independent and dependent variables. If an overall significant effect was detected, each MANOVA was followed up with univariate analyses of variance (ANOVA) and Tukey Honest Significant Difference (HSD) analyses for the continuous life satisfaction variable and logistic regression analyses for the ordinal SRH variable to determine which variables were accounting for the significant differences. Lastly, a MANOVA was also performed to test whether socioeconomic status (SES, as measured by free/reduced priced lunch) was significantly related to the independent physical activity variables.

Separate analyses were performed for the dependent variables (i.e. life satisfaction and SRH) for both males and females, owing to observed differences between males and females in previous physical activity research (Trost et al. 2002). Effect size (ES) was also calculated from the post-hoc comparisons for the life satisfaction variable to determine the magnitude of specific comparisons by dividing the mean difference by the pooled standard deviation. Effect size values provide an indication of the magnitude of observed differences and in a practical sense show the size of differences between means. Effect sizes of .20, .50, and .80 indicate small, medium, and large effects, respectively (Cohen 1988). Small effect sizes are generally not conceived as being practically important whereas medium and large effects are believed to be important.


MANOVA Results

Results indicated a significant overall difference between life satisfaction and SRH and the dependent study variables (F (4,241) = 9.16, p < .0001 and F (4,241) = 8.95, p < .0001, respectively). In addition, a significant overall difference between SES and the dependent study variables was also detected (F (4,241) = 2.44, p < .05). Therefore, all subsequent analyses were adjusted for SES.

ANOVA and Logistic Regression Results

Vigorous Exercise (Past 7 Days)

No significant differences were detected for males between life satisfaction and engaging in vigorous physical activity (F (2,107) = 2.99, p = .08) or between SRH and engaging in vigorous physical activity (OR = 5.91, CI = 0.91–38.31). However, significantly reduced life satisfaction was detected for females who reported not engaging in vigorous physical activity during the past 7 days (F (2,134) = 7.41, p < .01) (M = 4.36, SD = 0.67) when compared to females who reported engaging in vigorous physical activity during the past 7 days (M = 5.43, SD = 1.31) (ES = .75), but not between SRH and engaging vigorous physical activity (OR = 2.89, CI = 0.53–15.61)

TV Watching (Average School Day)

No significant differences were detected for males or females between television watching and satisfaction with life (F (2,107) = 0.11, p = .74 and F (2,133) = 0.16, p = .69, respectively) or SRH (OR = 2.42 CI: 0.29–20.20 and OR = 0.51 CI: 0.14–1.85, respectively).

Physical Education (PE) Class Attendance (Average School Week)

Similarly, no significant differences were detected for males or females between attendance in physical education classes and satisfaction with life (F (2,107) = 0.20, p = .66 and F (2,134) = 1.09, p = .30, respectively) or SRH (OR = 0.91 CI: 0.24–3.46 and OR = 2.31 CI: 0.64–8.30, respectively).

Sports Team Participation

Significantly reduced life satisfaction was detected for males who reported not playing on a sports team (F (2,107) = 8.95, p < .001) (M = 4.97, SD = 1.86) when compared to males who reported playing on a sports team (M = 5.90, SD = 1.27) (ES = .66). Significantly reduced life satisfaction was also detected for females who reported not playing on a sports team (F (2,127) = 27.83, p < .0001) (M = 4.59, SD = 1.34) when compared to females who reported playing on a sports team (M = 5.73, SD = 1.07) (ES = .80).

Furthermore, the odds of reporting fair/poor SRH significantly increased for males who reported not playing on a sports team when compared to males who reported playing on a sports team (OR = 5.40, CI = 1.30–22.39, p < .05). In addition, this same trend was also observed for females (OR = 30.92, CI = 3.74–255.43, p < .001).


This was the first study to investigate physical activity behaviors, life satisfaction, and self-rated health concurrently among middle school students. Although most middle school students in this study reported being mostly satisfied with their lives and good or better SRH, females in this study reported significantly reduced life satisfaction when not participating in vigorous exercise in the past 7 days. However, this same association was not seen for males. Notably, both of these findings stand at opposition with what has been observed with older high school adolescent populations (Valois et al. 2004).

It is possible that middle school aged females may view exercise and physical activity as more important contributors to health and well-being as compared to males at this age. For example, Robbins and colleagues (2009) found that 6th–8th grade females’ top reported benefit of physical activity was “to take care of myself, stay in shape, and be healthier” as compared to males’ top reported benefit of “improving or having more athletic skill.” However, the greatest reductions in adolescent physical activity also occur between 6th and 9th grade (Agazi et al. 2010; Caspersen et al. 2000). Although speculative, this reduction in vigorous physical activity in females may correspond to the activity itself becoming less important in the lives females in a direct sense, but more important to males as they mature.

Nevertheless, of potentially greater importance from this study are the observed associations between life satisfaction and SRH with sports participation for both males and females. Sports team participation may enhance school connectedness, social support, and bonding among friends and teammates, which are consistent with intrinsic values development. For example, McHale and colleagues (2005) found that middle school students who participated in sports had higher self-esteem and were viewed by teachers as being more socially competent than students not participating in sports. Kirkcaldy et al. (2002) found adolescent sports involvement to be associated with improved psychological well-being, including reduced anxiety and depression and fewer social problems. In addition, Babiss and Gangwisch (2009) found that as sports participation increased, the odds of depression symptoms and suicidal ideation among adolescents significantly decreased. Increased self-esteem and social support from the sports participation may potentially contribute to this relationship (Babiss and Gangwisch 2009). Finally, sports participation has been shown to have significant and positive effects on self-reported happiness (Rasciute and Downward 2010) and life satisfaction among older adolescents (Valois et al. 2004). Thus, the current study contributes to the literature by showing the benefits of youth sport participation on SRH and life satisfaction among even younger youth at a critical juncture in adolescent development.

Lastly, television watching was found to be unrelated to middle school student life satisfaction and SRH in this study. Although increased television watching has been used as proxy measure of sedentary behavior and linked to lower levels of physical activity, increased weight, and diabetes (Eisenmann et al. 2002; Hu et al. 2001; Anderson et al. 1998), recent meta-analytic findings suggest the relationship likely carries limited clinical significance (Marshall et al. 2004). Results reported here could be reflective of this finding.


The MSYRBS relies on self-reports to obtain measurements of physical activity, SRH, and life satisfaction. Previous research on using self-reports to collect objective data and Kappas of the MSYRBS have shown moderate reliability and validity (Palmer 2002; Zullig et al. 2006). However, factors like recollection of events, social desirability, and questionnaire administration in a peer-group setting have the potential to influence validity of self-reported data (Fan et al. 2006). For example, social desirability may persuade younger students, especially, to overestimate traits that are deemed desirable while reporting underestimates of undesirable health-related traits (Klesges et al. 2004).

A second limitation is the convenience sample used for this study, which included an over-representation of students reporting “white” as their ethnicity, as well as an over-representation of students in 7th grade, which may limit the generalizability of the results. However, the validity of this sample’s self reports is supported by national high school data collected by Eaton et al. (2006). Still, this sample may not be considered representative of the national geographic population particularly with ethnicity and age characteristics.

Finally, a cross-sectional study design does not imply causality. However, preliminary relationships between physical activity and positive health outcomes have now been confirmed among middle school students and continue through young adulthood. These findings suggest intervention efforts may be best conducted in the middle school years or earlier for the greatest impact, especially given the significant decline of physical activity observed among middle school adolescents (Agazi et al. 2010; Caspersen et al. 2000).

Conclusions and Implications

Despite the benefits, studies have shown that middle school students report barriers to physical activity including being too tired, being too busy, and reporting aches and pains from physical activity (Robbins et al. 2009). Given the findings from this study in combination with previous research, it will be important for future studies to determine ways to reduce potential barriers and increase participation in physical activity at young ages. For example, Sallis and colleagues (2001) found engagement in school-time physical activity outside of physical education class was higher for male and female middle school students when there was more adult supervision and when physical activity areas had been improved, such as through the addition of sports equipment. Previous interventions have also been successful in increasing adolescent physical activity when targeting physical activity during breaks at school, such as lunch and recess, and during summer day camps (Jago and Baranowski 2004; Stellino et al. 2010).

Other approaches schools have taken are structural changes in the school day that allowed for greater student physical activity participation and increased positive interactions between students and teachers. Such environmental interventions have resulted in measured increases in the percentage of students passing proficiency exams in writing, reading, and math while also observing concurrent reductions in nurse visits and discipline referrals with little disruption to classroom instruction time (Sibley et al. 2008).

It will also be important for organized programs emphasizing life-long physical activity to be promoted in schools and community organizations. Schools should deliver comprehensive health and physical education programs that are developmentally appropriate, and provide youth with enjoyable experiences that build exercise self-efficacy, provide sufficient levels of physical activity, and promote cognitive learning related to lifelong participation. For instance, Straus and colleagues (2001) found that high-level physical activity, such as that found in sports participation, was positively associated with increased self-efficacy, as well as increased self-esteem, for youth ages 10–16. Notably, positive affect is related to increased adolescent satisfaction with life (Huebner and Dew 1996), and increased adolescent life satisfaction is related to greater SRH (Zullig et al. 2005). Thus, understanding the mechanisms by which physical activity and sports participation may be associated with improved life satisfaction and self-rated health will determine best practices for positively impacting the health and quality of life of middle school students.

Copyright information

© Springer Science+Business Media B.V./The International Society for Quality-of-Life Studies (ISQOLS) 2010