Skip to main content
NCBI home page
Journal of Pediatric Psychology logoLink to Journal of Pediatric Psychology
. 2013 Jan 31;38(4):387–397. doi: 10.1093/jpepsy/jss178

Weight Status as a Moderator of the Relationship Between Motivation, Emotional Social Support, and Physical Activity in Underserved Adolescents

Sara M St George 1,✉, Dawn K Wilson 1, Hannah G Lawman 1, M Lee Van Horn 1
PMCID: PMC3695650  PMID: 23378172

Abstract

Objective This study examined weight status as a moderator of the relationship between motivation (controlled, autonomous, regulatory), emotional social support (parents, peers) and moderate-to-vigorous physical activity (MVPA) in underserved adolescents (ethnic minority, low-income). Methods Participants from the Active by Choice Today Trial (n = 1,416; 54% girls, 73% African American, 52% overweight/obese) completed baseline measures, including height and weight, psychosocial surveys, and 7-day accelerometry estimates. Weight status was defined by body mass index z-score (zBMI). Results Weight status moderated the effects of controlled, autonomous, and regulatory motivation on MVPA, such that these variables were more strongly associated with MVPA in adolescents with lower versus higher zBMI scores. Conclusions A better understanding of why motivation is not related to MVPA in underserved youth with a higher weight status is needed. Future pediatric obesity treatment in underserved youth may need to move beyond motivation into environmental factors associated with long-term behavior change.

Keywords: adolescents, motivation, physical activity, social support, weight

Introduction

Adolescents from underserved backgrounds (low-income, ethnic minority) have among the highest rates of obesity (Ogden, Carroll, Kit, & Flegal, 2012). Although engaging in physical activity (PA) has been shown to improve various pediatric outcomes (Janssen & LeBlanc, 2010), <10% of adolescents engage in 60 min of daily moderate-to-vigorous PA (MVPA) (Troiano et al., 2008). Low income, ethnic minority, and youth with a higher weight status are also less active than those of a higher income, nonminority, or a lower weight status (Delva, Johnston, & O'Malley, 2007; Janssen et al., 2005). Because psychosocial variables such as motivation and emotional social support (i.e., esteem-enhancing support) have been associated with sustained PA behavior (Teixeira, Carraça, Markland, Silva, & Ryanet al., 2012), developing a better understanding of these factors in underserved youth is essential to the promotion of PA, prevention of obesity, and overall decrease in health disparities. This study examined the relationship between motivation (controlled, autonomous, regulatory), emotional support (parents, peers), and MVPA by weight status in a sample of primarily African American sixth graders to determine how these variables may be more effectively integrated into obesity treatment and prevention efforts.

Motivational frameworks (e.g., Self Determination Theory [SDT; Ryan & Deci, 2000]) have been used to understand youth PA behavior (Motl, 2007). According to SDT, motivation rests along a continuum ranging from extrinsic, or more controlled forms of motivation (i.e., engaging in a behavior to satisfy external demands), to intrinsic, or more autonomous forms of motivation (i.e., engaging in a behavior for its inherent satisfaction). Experiencing inherent satisfaction from engaging in PA may also be linked with increased regulatory motivation or a willingness to incorporate PA it into one’s daily routine (Wilson et al., 2002, 2005). Although autonomous and regulatory motivation are both forms of intrinsic motivation, regulatory motivation additionally reflects ongoing behavioral regulation of PA. Overall, SDT suggests intrinsically motivated behavior changes will be sustained longer than those driven by extrinsic factors (Ryan & Deci, 2000). Furthermore, social factors, such as emotional support create the context for facilitating intrinsic motivation by supporting an individual’s needs for autonomy (feeling of choice and control), competence (feeling one has the skills to engage in a behavior), and belongingness (feeling valued). Although a well-documented positive relationship exists between motivation and youth PA participation (Cox, Smith, & Williams, 2008; Lawman, Wilson, Van Horn, Resnicow, & Kitzman-Ulrich, 2011; Wilson, Mack, & Grattan, 2008), this relationship has not been examined by weight status in underserved youth.

Youth who are either obese or perceive themselves to be overweight have been shown to be less autonomously motivated to engage in PA (Ingledew & Sullivan, 2002; Power, Ullrich-French, Steele, Daratha, & Bindler, 2011), are more likely to drop out from exercise (Gillison, Standage, & Skevington, 2011), and have endorsed more extrinsic exercise motives than normal-weight youth (Gillison, Standage, & Skevington, 2006). One study found that body mass index (BMI) was negatively associated with intrinsic exercise motives (e.g., enjoyment), and positively associated with extrinsic exercise motives (e.g., weight management) in adolescents (Ingledew & Sullivan, 2002). Differences in the relationship between motivation type and weight status have been studied primarily in Caucasian youth and are understudied in ethnically diverse samples. One study examining PA enjoyment found no difference by weight status in African American adolescent girls (Ward et al., 2006). Although a significant positive relationship between intrinsic motivation and MVPA has been previously established in the present sample (Lawman et al., 2011), neither extrinsic motivation nor differences across motivation type by weight status have been examined. Overall, previous studies have used either small primarily Caucasian samples or self-reported PA as the primary outcome, and many have failed to examine a range of motivational and social contextual variables together to understand PA in ethnically diverse youth of varying weight statuses. To determine how best to increase PA adherence in underserved youth across the weight spectrum, the present study explores the effect of motivation, emotional support, and adolescent weight status on objectively assessed MVPA.

Numerous studies have indicated youth with a higher weight status are more likely than normal-weight youth to experience negative social interactions (Janssen, Craig, Boyce, & Pickett, 2004; Neumark-Sztainer et al., 2002), which have been inversely associated with youth PA levels (Storch et al., 2006). Although emotional social support or the provision of caring, empathy, love, and trust (House, 1981; e.g., through encouragement from parents and peers) has been positively associated with adolescent PA (Biddle & Goudas, 1996; Dowda, Dishman, Pfeiffer, & Pate, 2007; Martin & McCaughtry, 2008; Springer, Kelder, & Hoelscher, 2006; Trost et al., 2003), findings on how these relationships vary as a function of adolescent weight status are somewhat mixed. While some studies have found no differences in parental and peer support for PA by youth weight status (Trost, Kerr, Ward, & Pate, 2001; Ward et al., 2006), others have shown positive associations with PA in normal-weight but not overweight youth (De Bourdeaudhuij et al., 2005; Kitzman-Ulrich, Wilson, Van Horn, & Lawman, 2010). Given parental and peer emotional support are often targeted in pediatric interventions, a better understanding of how these variables differ by weight status is critical to informing future obesity interventions in underserved adolescents.

The purpose of this study is to examine the interaction between a range of motivational factors, social support, and weight status in predicting MVPA in underserved adolescents. Based on previous studies in primarily Caucasian youth, it was hypothesized that autonomous motivation, regulatory motivation, and emotional support (parents, peers) would each have an independent positive association with MVPA and be stronger predictors of MVPA in youth with a lower versus higher weight status. Furthermore, for controlled motivation, it was hypothesized that although this variable would also have an independent positive association with MVPA, it would be a stronger predictor of MVPA in those with a higher versus lower weight status.

Method

Participants

Participants were sixth-grade adolescents who completed baseline measures as part of their participation in a randomized school-based trial (“Active by Choice Today”; ACT) examining a motivational plus behavioral skills intervention designed to increase PA (Wilson et al., 2008, 2011). Sixth graders were included given their unique developmental transition into middle school and thus increased vulnerability for declines in PA. While findings of the larger trial demonstrated an intervention effect during the afterschool program, this study expands on the original trial by examining how body mass index z-score (zBMI) may moderate the effects of motivation and support on baseline MVPA. Participants were recruited from schools classified as underserved (i.e., ≥50% ethnic minority, ≥50% free/reduced cost lunch) through presentations at orientations, homeroom visits, and pep rallies. Adolescents were eligible to participate if they were enrolled in sixth grade, obtained parental consent, agreed to study participation, and were available for a 6-month follow-up. Adolescents were excluded if they had a medical condition that prevented PA participation, were developmentally delayed such that intervention materials were not cognitively appropriate, or were in treatment for a psychiatric disorder. Approval from the University of South Carolina Institutional Review Board was obtained before beginning the trial. Parental informed consent and adolescent assent were obtained before data collection.

Procedure

Before randomization and the start of the intervention, an independent team of trained measurement staff collected baseline measures, including demographics, anthropometric data (height, weight, waist circumference), psychosocial surveys, and 7-day accelerometry estimates of MVPA. All participants received a $5 gift card incentive for completing baseline measures.

Measures

Anthropometric Measures

Height was measured using a Shorr Height Measuring Board, and weight was measured with a SECA 880 digital scale. Two measures of height and weight were taken, and the average score was used in anthropometric calculations. BMI percentiles and zBMI were calculated with EpiInfo (Version 3.5.1) using the most recently available Centers for Disease Control and Prevention growth reference curves (Kuczmarski et al., 2002), which standardizes data to population values.

Motivational Factors Related to PA

In a previously published study, psychometrics of the original motivational measures used in the ACT trial were examined (Lawman et al., 2011). In an effort to develop concise measures for assessing a range of motivational factors, a more recent study examining the factor structure of these motivational measures revealed that a three-factor model of motivation (i.e., controlled, autonomous, regulatory motivation), provided an adequate fit to the data [χ2 (87) = 706.2, p < .01, CFI = .95, TLI = .94, RMSEA = .07] (Lawman, Wilson, Van Horn, & Resnicow, 2013). These revised scales are briefly described below.

Controlled and Autonomous Motivation for PA

Controlled motivation (i.e., engaging in behavior to gain reward or avoid punishment) and autonomous motivation (i.e., engaging in behavior for its inherent satisfaction) were assessed using a modified version of the Exercise Self Regulation Questionnaire (SRE-Q) (Ryan & Deci, 2000; Ryan & Connell, 1989). As used in the ACT trial, the SRE-Q was composed of 13 items, which assessed intrinsic motivation, identified regulation, introjected regulation, and extrinsic motivation. Within the more recent framework, several introjected and extrinsic items were conceptualized as controlled motivation while intrinsic and identified items were conceptualized as autonomous motivation (Deci & Ryan, 2008). Items were standardized and averaged to create separate measures of controlled (five items) and autonomous motivation (five items). Participants responded on a 3-point scale (1 = “Not like me,” 2 = “A little like me,” 3 = “A lot like me”) to items such as “I am active so I can stay in shape” (controlled motivation) and “I am active because I enjoy it” (autonomous motivation). Alpha reliability of the controlled (α = .70) and autonomous motivation scales (α = .77) were acceptable, and to demonstrate construct validity, the measures were correlated with MVPA (r = .11, p < .01 controlled scale; r = .09, p < .01 autonomous scale).

Regulatory Motivation for PA

Regulatory motivation for PA (i.e., willingness to incorporate PA into one’s daily routine) was measured using a scale originally developed by Wilson and colleagues (2002, 2005) and later modified (Lawman et al., 2011, 2013). Modifications were made to the number of items. Although the original scale contained 10 items (Wilson et al., 2002, 2005), five items were standardized and averaged in this study to create one measure of regulatory motivation. Participants responded on a 3-point scale (1 = “Not like me,” 3 = “A lot like me”) to items such as, “I plan how I can be active every day.” Alpha reliability of this revised scale (α = .82) was acceptable, and the measure was correlated with enjoyment for PA (r = .60, p < .01) and MVPA (r = .17, p < .01).

Parent and Peer Emotional Support for PA

Parent and peer emotional support for PA was measured using a modified version of the Support for Exercise Scales (Sallis, Grossman, Pinski, Patterson, & Nader, 1987). Twelve items were standardized and averaged to create a measure of parent (six items) and peer emotional support (six items). Participants responded on a 3-point scale (1 = “None,” 2 = “A few times,” 3 = “Many times”) to items such as, “In the past month, how often has [your parent or other adult, a friend] told you to stick with being active?” Alpha reliability of the parent (α = .80) and peer emotional support scales (α = .79) was acceptable, and the measures were correlated with self-efficacy for PA (r = .26, p < .01 parent scale; r = .42 peer scale) and MVPA (r = .05, p < .01 parent scale; r = .13, p < .01 peer scale).

Moderate-to-Vigorous PA

MVPA was assessed with a 7-day estimate using Actical omni-directional accelerometers (Mini-Mitter, Bend, Oregon). Actical has been previously validated as a measure of youth PA (Puyau, Adolph, Vohra, Zakeri, & Butte, 2004). Each day of Actical data is divided into five time intervals (6–9 a.m., 9 a.m.–2 p.m., 2–5 p.m., 5–8 p.m., and 8 p.m.–12 a.m.) to allow for examination of PA by specific time periods throughout the day (Catellier et al., 2005). Data were recorded in 1-min epochs (Welk, Schaben, & Morrow, 2004), and raw activity data was converted into metabolic equivalents (METS; moderate PA = 3–5.9 METS, vigorous PA = 6–8.9, and MVPA = 3–8.9 METS) using previously validated cutoff estimates (Puyau et al., 2004). Twenty consecutive zero counts were used to indicate nonwear. Missing data for a given participant were identified by periods during which the Actical was worn for <80% of the time when at least 70% of participants wore the monitor. Overall, 37% of accelerometry data were missing due to nonwear, lost accelerometers, or accelerometer malfunction; only 3% of adolescents were missing all baseline accelerometry data. Missing data were dealt with using multiple imputation procedures (see data analyses) (Catellier et al., 2005). Following imputation, minutes of MVPA were summed for each interval to produce daily counts of MVPA, and then 7 days were averaged to provide one measure of average daily MVPA.

Data Analyses

Although the original sample included 1,422 adolescents, two outliers (495.78 min/day of MVPA, −8.9 zBMI; both ≥8 standard deviations above or below their respective means) were removed. The final sample size was 1,420 adolescents. Although this study contains a nested design (adolescents within schools), it focuses on individual-level processes. The outcome variables have low intraclass correlation coefficients (0.01–0.02) justifying an individual level of analysis. In addition, the design effects, or the multiplier by which standard errors are increased due to clustering, were low (1.006–1.012) (Neuhaus & Segal, 1993).

Multiple imputations (m = 40) were conducted with the statistical package R (Version 2.15.2) using the package “pan,” which models multilevel imputations and assumes a normal distribution. Results from the procedure were subsequently combined and produced an estimate of the average R-squared and F values across imputations. Three ordinary least squares (OLS) regression models were used to test the moderating effects of zBMI on the relationship between motivational factors, social factors, and MVPA. Given moderate positive correlations between the three motivational variables (r = .68 for autonomous and regulatory motivation, r = .57 for controlled and autonomous motivation; r = .51 for controlled and regulatory motivation), models were run separately for these variables to avoid potential multicollinearity. Variables were either contrast coded (sex, race), centered (zBMI), or standardized (motivational and support variables) to facilitate model interpretation, with the intercept representing the mean minutes per day of MVPA for the average adolescent. Although zBMI was already standardized based on population values, it was centered in the present dataset by subtracting the mean of zBMI to reduce multicollinearity. Owing to differences in adolescent PA by sex and race, all models controlled for these variables. An example of the regression equation follows: MVPA = β0 + β1 sex + β2 race + β3 zBMI + β4 regulatory motivation + β5 parent support + β6 peer support + β7 zBMI*regulatory motivation + β8 zBMI*parent support + β9 zBMI*peer support. Finally, post hoc tests of significant interactions were conducted to determine significance of simple slopes (i.e., whether or not slopes differed significantly from zero).

Results

Participant Demographic & Psychosocial Characteristics

For descriptive purposes, demographic characteristics are presented in Table I by BMI percentile category (<85th percentile = normal weight, 85th–95th percentile = overweight, >95th percentile = obese). One-way analysis of variances (ANOVAs) revealed significant differences between normal, overweight, and obese adolescents on age [F(1, 1418) = 3.96, p < .05] and MVPA [F(1, 1418) = 41.57, p < .01]. Follow-up t tests revealed adolescents in the normal-weight category were older than those in the overweight category (t = 2.29, df = 934, p < .05) and participated in significantly higher levels of MVPA than those in the obese category (t = 6.70, df = 1,166, p < .01). Similarly, adolescents in the overweight category had significantly greater MVPA levels (t = 4.01, df = 734, p < .01) than those in the obese category. There were no significant differences between groups on motivational variables, parent support, or peer support.

Table I.

Participant Characteristics by BMI Percentile Category (Normal Weight, Overweight, Obese)

Variable Normal weight (<85th percentile) Overweight (85th–95th percentile) Obese (>95th percentile) Total
N 684 (48%) 252 (18%) 484 (34%) 1,420
Sex
    Girls 363 (47%) 127 (16%) 281 (36%) 771 (54%)
    Boys 321 (50%) 125 (19%) 203 (31%) 649 (46%)
Race
    African American 498 (48%) 180 (17%) 362 (35%) 1,040 (73%)
    Caucasian 186 (49%) 72 (19%) 122 (32%) 380 (27%)
Age (years)*,a 11.38 (0.60) 11.28 (0.57) 11.32 (0.57) 11.34 (0.59)
Waist circumference (cm)**,a,b,c 64.38 (9.48) 71.19 (7.08) 80.22 (12.30) 70.99 (12.40)
MVPA (min/day)**,b,c 46.72 (27.76) 44.25 (29.04) 36.51 (22.39) 42.80 (26.68)
Self-efficacy 2.33 (0.40) 2.35 (0.39) 2.29 (0.39) 2.32 (0.40)
Parent emotional support for PA 1.86 (0.52) 1.93 (0.52) 1.94 (0.52) 1.90 (0.52)
Peer emotional support for PA 1.91 (0.53) 1.97 (0.54) 1.90 (0.53) 1.92 (0.53)
Controlled motivation for PA 2.24 (0.51) 2.33 (0.49) 2.28 (0.48) 2.27 (0.50)
Autonomous motivation for PA 2.51 (0.47) 2.56 (0.44) 2.54 (0.43) 2.53 (0.45)
Regulatory motivation for PA 2.29 (0.55) 2.33 (0.50) 2.28 (0.53) 2.30 (0.53)
Home equipment for PA**,b,***,c 2.71 (0.53) 2.70 (0.52) 2.62 (0.57) 2.68 (0.54)
Open in a new tab

Note. Values are expressed as frequencies (%) or means (standard deviations).

MVPA = moderate-to-vigorous physical activity; PA = physical activity.

aSignificant difference between normal and overweight groups.

bSignificant difference between normal and obese groups.

cSignificant difference between overweight and obese groups.

*p < .05, **p < .01, ***p < .10.

Regression Analyses

Effects of zBMI, Controlled Motivation, and Social Support on MVPA

The model including sex, race, zBMI, controlled motivation, parent support, peer support, and interactions between zBMI and controlled motivation, parent support, and peer support significantly predicted adolescent MVPA [F(9, 1410) = 25.55, p < .01] and accounted for 14% of the variance in MVPA (See Table II). Significant main effects were found for sex, zBMI, and peer support predicting MVPA, such that boys, those with a lower weight status, and those with greater perceived peer social support were more physically active than girls, those with a higher weight status, and those with lower perceived peer social support, respectively. Weight status (zBMI) was a significant moderator of the relationship between controlled motivation and MVPA (B = −2.01, SE = 1.05, t = −1.92, p = .05), such that controlled motivation was more strongly associated with MVPA in adolescents with lower versus higher zBMI scores (see Figure 1, Graph A). At one standard deviation above the mean for controlled motivation, the predicted MVPA value was 52.61 min/day for youth with low zBMI and 39.19 min/day for youth with high zBMI (a difference of 13.42 min/day of MVPA). Follow-up analyses to examine the significant interaction were conducted, and the calculation of simple slopes indicated that the simple slope for adolescents with low (slope = 3.63, SE = 1.39, t = 2.60, df = 1,416, p < .01) but not high zBMI was significantly different from zero, suggesting controlled motivation was positively associated with MVPA at low zBMI levels only. No other main effects or interactions were significant.

Table II.

Summary of Regression Analyses for Variables Predicting MVPA (n = 1420)

Unstandardized coefficients
 Standardized coefficients
Variables B SE 95% Confidence interval β t FMI
Controlled motivation model
Overall model F = 25.55 (9, 1410), p < .01, R2 = .14
    Intercept 44.04 0.90 42.27 < B0 < 45.82 .01 48.67** .28
    Sex −15.3 1.48 −18.20 < B1 < −12.40 −.28 −10.35** .13
    Race 0.75 1.72 −2.63 < B2 < 4.13 −.17 0.58 .2
    zBMI −4.7 0.69 −6.05 < B3 < −3.35 .01 −6.81** .19
    Controlled motivation 1.85 1.2 −0.51 < B4 < 4.22 .07 1.54 .17
    Emotional parent support 0.23 1.28 −2.28 < B5 < 2.73 .01 0.18 .14
    Emotional peer support 3.33 1.27 0.84 < B6 < 5.82 .12 2.62** .13
    zBMI × controlled motivation −2.01 1.05 −4.07 < B7 < 0.04 −.07 −1.92*** .13
    zBMI × emotional parent support −0.47 1.17 −2.77 < B8 < 1.82 −.02 −0.4 .11
    zBMI × emotional peer support −0.61 1.19 −2.94 < B9 < 1.72 −.02 −0.51 .1
Autonomous motivation model
Overall model F = 25.69 (9, 1410), p < .01, R2 = .14
    Intercept 43.96 0.90 42.19 < B0 < 45.73 .01 48.74** .28
    Sex −15.68 1.47 −18.56 < B1 < −12.80 −.29 −10.67** .13
    Race 1.07 1.71 −2.29 < B2 < 4.43 −.17 0.62 .2
    zBMI −4.61 0.68 −5.95 < B3 < −3.27 .02 −6.74** .19
    Autonomous motivation 1.46 1.15 −0.80 < B4 < 3.72 .05 1.27 .22
    Emotional parent support 0.27 1.27 −2.22 < B5 < 2.76 .01 0.22 .14
    Emotional peer support 3.3 1.28 0.78 < B6 < 5.81 .12 2.57* .13
    zBMI × autonomous motivation −2.15 0.99 −4.09 < B7 < −0.22 −.08 −2.19* .17
    zBMI × emotional parent support −0.58 1.16 −2.85 < B8 < 1.70 −.02 −0.5 .12
zBMI × emotional peer support −0.39 1.2 −2.75 < B9 < 1.96 −.01 −0.33 .1
Regulatory motivation model
Overall model F = 27.09 (9, 1410), p < .01, R2 = .15
    Intercept 43.79 0.9 42.03 < B0 < 45.56 .0 48.66** .28
    Sex −15.08 1.47 −17.96 < B1 < −12.21 −.28 −10.29** .13
    Race 1.24 1.71 −2.11 < B2 < 4.59 −.16 0.72 .2
    zBMI −4.5 0.68 −5.83 < B3 < −3.16 .02 −6.62** .19
    Regulatory motivation 3.58 1.14 1.34 < B4 < 5.82 .13 3.14** .19
    Emotional parent support −0.11 1.27 −2.59 < B5 < 2.38 .0 −0.08 .15
    Emotional peer support 2.18 1.33 −0.43 < B6 < 4.78 .08 1.64 .13
    zBMI × regulatory motivation −2.42 0.99 −4.36 < B7 < −0.48 −.09 −2.44* .15
    zBMI × emotional parent support −0.46 1.16 −2.74 < B8 < 1.83 −.02 −0.39 .13
    zBMI × emotional peer support 0.01 1.23 −2.39 < B9 < 2.41 0 0.01 .09
Open in a new tab

Note. FMI = fraction of missing information; SE = standard error; zBMI = body mass index z-score.

*p < .05, **p < .01, ***p < .10.

Figure 1.

Figure 1.

Open in a new tab

Interaction of zBMI and (A) controlled motivation, (B) autonomous motivation, and (C) regulatory motivation predicting minutes per day of MVPA. High and low values represent 1 standard deviation above and below the mean.

Effects of zBMI, Autonomous Motivation, and Social Support on MVPA

The overall model significantly predicted adolescent MVPA [F(9, 1410) = 25.69, p < .01] and accounted for 14% of the variance in MVPA. Significant main effects were found for sex, zBMI, and peer support predicting MVPA, such that boys, those with a lower weight status, and those with greater perceived peer social support were more physically active than girls, those with a higher weight status, and those with lower perceived peer social support, respectively. Weight status significantly moderated the relationship between controlled motivation and MVPA (B = −2.15, SE = 0.99, t = −2.19, p < .05), such that autonomous motivation was more strongly associated with MVPA in adolescents with lower versus higher zBMI scores (see Figure 1, Graph B). At one standard deviation above the mean for autonomous motivation, the predicted MVPA value was 52.19 min/day for youth with low zBMI and 38.66 min/day for youth with high zBMI (a difference of 13.53 min/day of MVPA). The calculation of simple slopes for the significant interaction between zBMI and autonomous motivation indicated that the simple slope for adolescents with low (slope = 3.49, SE = 1.26, t = 2.78, df = 1,416, p < .01) but not high zBMI was significantly different from zero, suggesting autonomous motivation was positively associated with MVPA at low zBMI levels only. No other main effects or interactions were significant.

Effects of zBMI, Regulatory Motivation, and Social Support on MVPA

The overall model significantly predicted adolescent MVPA [F(9, 1410) = 27.09, p < .01] and accounted for 15% of the variance in MVPA. Significant main effects were found for sex, zBMI, and regulatory motivation, such that boys, those with a lower weight status, and those with greater regulatory motivation were more physically active than girls, those with a higher weight status, and those with lower regulatory motivation, respectively. Weight status significantly moderated the relationship between regulatory motivation and MVPA (B = −2.42, SE = 0.99, t = −2.44, p < .05), such that regulatory motivation was more strongly associated with MVPA in adolescents with lower versus higher zBMI scores (see Figure 1, Graph C). At one standard deviation above the mean for regulatory motivation, the predicted MVPA value was 54.29 min/day for youth with low zBMI and 40.46 min/day for youth with high zBMI (a difference of 13.83 min/day of MVPA). Follow-up analyses on the significant interaction indicated that the simple slope for adolescents with low (slope = 6.32, SE = 1.33, t = 4.76, df = 1,416, p < .01) but not high zBMI was significantly different from zero, suggesting regulatory motivation was positively associated with MVPA at low zBMI levels only. No other main effects or interactions were significant.

Home PA Equipment by Weight Status

Follow-up analyses were conducted to examine differences by weight status in a single item measuring availability of home PA equipment. A one-way ANOVA revealed significant between-group differences [F(1, 1418) = 7.07, p < .01], such that adolescents in the normal-weight versus obese category reported having greater access to home PA equipment (t = 2.77, df = 1,166, p < .01).

Discussion

The purpose of the present study was to evaluate how motivation and social support would differentially predict MVPA in underserved youth of varying weight statuses. The use of a large sample size, examination of both motivational and social variables in underserved, predominantly African American adolescents, the use of accelerometers to measure MVPA, and multiple imputation methods to account for missing data were among the strengths of the study. Findings indicated that weight status moderated the effects of motivational variables, including controlled, autonomous, and regulatory motivation on MVPA. As hypothesized, underserved youth with a lower weight status displayed greater overall daily minutes of MVPA when they had higher levels of autonomous and regulatory motivation; however, contrary to what was expected, this association was also true for controlled motivation. Further examination of the simple slopes for these interactions indicated that controlled, autonomous, and regulatory motivation were positively associated with MVPA in youth with a low (but not a high) weight status. Overall, it was estimated that youth with high levels of controlled, autonomous, and regulatory motivation and a lower weight status would engage in ∼13–14 more daily min of MVPA compared with youth with similar levels of motivation but a higher weight status. This difference in MVPA is clinically meaningful in that even 5–15 min bouts of PA can improve health outcomes (Thomas et al., 2009; Wen et al., 2011). Across the models tested, consistent main effects were also found for sex and zBMI, such that boys and adolescents with a lower zBMI were significantly more physically active than girls and adolescents with lower zBMI. For all but the regulatory motivation model, main effects of peer emotional support also emerged, such that adolescents with greater emotional support from peers were more active than those with lower support. Parent support was not a significant predictor of MVPA, and the relationship between social support variables (parent, peers) and MVPA did not vary based on adolescent weight status.

Interestingly, the three motivational variables (controlled, autonomous, and regulatory) were the only psychosocial variables to interact with adolescent weight status in predicting youth MVPA. Previous studies comparing psychosocial correlates of PA (e.g., enjoyment, social support) in ethnically diverse adolescents have found no differences by weight status (Trost et al., 2001; Ward et al., 2006). The finding that parental and peer emotional support did not interact with weight status in predicting MVPA is consistent with these studies. De Bourdeaudhuij et al. (2005) concluded that specific tailoring on psychosocial correlates of PA was not necessary for overweight versus normal-weight adolescents, and findings from this study suggest this may be case for emotional support from parents and peers. Lack of findings related to parent emotional support may reflect a mismatch between type of parental support provided and adolescent support preferences; for example, qualitative research in African American adolescents has suggested youth desire greater amounts of tangible PA support (St. George & Wilson, 2012; Wright, Wilson, Griffin, & Evans, 2010). Thus, future studies should examine how other types of support, namely tangible support, interact with weight status in predicting MVPA (Beets, Cardinal, & Alderman, 2010). When it comes to motivation, however, this study expands on previous work by highlighting that motivation (regardless of type) does not seem to be working to predict MVPA in underserved youth with a higher weight status the way it works in their lower weight counterparts.

In fact, findings of the present study related to motivation have potentially important clinical implications for the prevention and treatment of pediatric obesity in underserved youth. Future prevention efforts aimed at underserved adolescents in the normal-weight range may increase their PA levels by targeting PA motivation. For obesity treatment efforts, however, merely increasing PA motivation in underserved youth who are already overweight or obese may not be an effective approach to increasing actual PA. Although the American Academy of Pediatrics has recommended motivational interviewing (MI) as an effective method to promote weight loss behaviors (Davis et al., 2007), this study suggests that for overweight or obese underserved youth, having higher levels of motivation may not equate to higher levels of MVPA. A recent pilot study using MI to target weight-related behaviors in overweight African American adolescents found that although participants in the MI intervention increased their intrinsic motivation, their activity decreased (MacDonell, Brogan, Naar-King, Ellis, & Marshall, 2011). Overall, a better understanding of why motivation is not related to MVPA in underserved youth with a higher weight status is needed. Researchers and health care practitioners involved in the treatment of pediatric obesity in underserved populations may need to consider other intrapersonal (e.g., embarrassment in ability or physical appearance, self-efficacy) and environmental factors associated with promoting long-term PA behavior changes.

Given the lack of resources in low-income underserved areas, it could be that the immediate social context is not as important as is the broader environment in terms of sustaining behavior. Follow-up analyses conducted as a part of this study indicated a significant difference between youth with a normal versus obese weight status on the availability of resources in the home for engaging in PA. This finding is consistent with previous research, which emphasizes the importance of the home environment and broader social context on influencing PA, particularly in youth living in impoverished areas (Ferreira et al., 2007). Future research in overweight underserved youth should continue to examine the adolescent social context beyond, or in conjunction with, the immediate context of peers into other broader environmental facilitators and barriers for PA, which may drive PA (Davison & Lawson, 2006; Gordon-Larsen, Nelson, Page, & Popkin, 2006).

Limitations of this study should be noted. First, causal inferences cannot be made given the cross sectional nature of the data. In addition, given the sample was composed exclusively of sixth graders, it is unclear how these results would generalize to younger or older adolescents. However, this study has provided insights into the nature of the relationship between a range of psychosocial factors and adolescent weight status.

In summary, this study highlights that motivational factors interacted with adolescent weight status in predicting MVPA while social support from parents and peers did not. To further our understanding of social contextual factors that may be key to the prevention and treatment of pediatric obesity, examining broader environmental correlates associated with long-term PA behavior may be a good direction for future research with ethnic minority, low-income youth who are overweight or obese.

Funding

This work was support by grants funded by the National Institutes of Child Health and Human Development (grant numbers R01 HD 045693 awarded to D.K.W. and F31 HD 066944 awarded to S.M.St.G. and D.K.W.).

Conflicts of interest: None declared.

References

  1. Beets M W, Cardinal B J, Alderman B L. Parental social support and the physical activity-related behaviors of youth: A review. Health Education & Behavior. 2010;37:621–644. doi: 10.1177/1090198110363884. [DOI] [PubMed] [Google Scholar]
  2. Biddle S, Goudas M. Analysis of children's physical activity and its association with adult encouragement and social cognitive variables. Journal of School Health. 1996;66:75–78. doi: 10.1111/j.1746-1561.1996.tb07914.x. [DOI] [PubMed] [Google Scholar]
  3. Catellier D J, Hannan P J, Murray D M, Addy C L, Conway T L, Song Y, Rice J C. Imputation of missing data when measuring physical activity by accelerometry. Medicine and Science in Sports and Exercise. 2005;37(11 Suppl):S555–S562. doi: 10.1249/01.mss.0000185651.59486.4e. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Cox A, Smith A, Williams L. Change in physical education motivation and physical activity behavior during middle school. Journal of Adolescent Health. 2008;43:506–513. doi: 10.1016/j.jadohealth.2008.04.020. [DOI] [PubMed] [Google Scholar]
  5. Davis M, Gance-Cleveland B, Hassink S, Johnson R, Paradis G, Resnicow K. Recommendations for prevention of childhood obesity. Pediatrics. 2007;120(Suppl):S229–S253. doi: 10.1542/peds.2007-2329E. [DOI] [PubMed] [Google Scholar]
  6. Davison K K, Lawson C T. Do attributes in the physical environment influence children's physical activity? A review of the literature. The International Journal Of Behavioral Nutrition And Physical Activity. 2006;3:19. doi: 10.1186/1479-5868-3-19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. De Bourdeaudhuij I, Lefevre J, Deforche B, Wijndaele K, Matton L, Philippaerts R. Physical activity and psychosocial correlates in normal weight and overweight 11 to 19 year olds. Obesity. 2005;13:1097–1105. doi: 10.1038/oby.2005.128. [DOI] [PubMed] [Google Scholar]
  8. Deci E L, Ryan R M. Facilitating optimal motivation and psychological well-being across life's domains. Canadian Psychology/Psychologie canadienne. 2008;49:14–23. [Google Scholar]
  9. Delva J, Johnston L D, O'Malley P M. The epidemiology of overweight and related lifestyle behaviors: Racial/ethnic and socioeconomic status differences among American youth. American Journal of Preventive Medicine. 2007;33:S178–S186. doi: 10.1016/j.amepre.2007.07.008. [DOI] [PubMed] [Google Scholar]
  10. Dowda M, Dishman R, Pfeiffer K, Pate R. Family support for physical activity in girls from 8th to 12th grade in South Carolina. Preventive Medicine. 2007;44:153–159. doi: 10.1016/j.ypmed.2006.10.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Ferreira I, van der Horst K, Wendel-Vos W, Kremers S, van Lenthe F J, Brug J. Environmental correlates of physical activity in youth—A review and update. Obesity Reviews. 2007;8:129–154. doi: 10.1111/j.1467-789X.2006.00264.x. [DOI] [PubMed] [Google Scholar]
  12. Gillison F, Standage M, Skevington S. Relationships among adolescents' weight perceptions, exercise goals, exercise motivation, quality of life and leisure-time exercise behaviour: A self-determination theory approach. Health Education Research. 2006;21:836–847. doi: 10.1093/her/cyl139. [DOI] [PubMed] [Google Scholar]
  13. Gillison F B, Standage M, Skevington S M. Motivation and body-related factors as discriminators of change in adolescents' exercise behavior profiles. Journal of Adolescent Health. 2011;48:44–51. doi: 10.1016/j.jadohealth.2010.05.006. [DOI] [PubMed] [Google Scholar]
  14. Gordon-Larsen P, Nelson M C, Page P, Popkin B M. Inequality in the built environment underlies key health disparities in physical activity and obesity. Pediatrics. 2006;117:417–424. doi: 10.1542/peds.2005-0058. [DOI] [PubMed] [Google Scholar]
  15. House J S. Work stress and social support. Reading, MA: Addison-Wesley; 1981. [Google Scholar]
  16. Ingledew D, Sullivan G. Effects of body mass and body image on exercise motives in adolescence. Psychology of Sport and Exercise. 2002;3:323–338. [Google Scholar]
  17. Janssen I, Craig W, Boyce W, Pickett W. Associations between overweight and obesity with bullying behaviors in school-aged children. Pediatrics. 2004;113:1187–1194. doi: 10.1542/peds.113.5.1187. [DOI] [PubMed] [Google Scholar]
  18. Janssen I, Katzmarzyk P, Boyce W, Vereecken C, Mulvihill C, Roberts C, Currie C, Pickett W Health Behaviour in School-Aged Children Obesity Working Group. (2005) Comparison of overweight and obesity prevalence in school aged youth from 34 countries and their relationships with physical activity and dietary patterns. Obesity Reviews. 6:123–132. doi: 10.1111/j.1467-789X.2005.00176.x. [DOI] [PubMed] [Google Scholar]
  19. Janssen I, LeBlanc A G. Systematic review of the health benefits of physical activity and fitness in school-aged children and youth. International Journal of Behavioral Nutrition and Physical Activity. 2010;7:40–55. doi: 10.1186/1479-5868-7-40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kitzman-Ulrich H, Wilson D, Van Horn M, Lawman H. Relationship of body mass index and psychosocial factors on physical activity in underserved adolescent boys and girls. Health Psychology. 2010;29:506–513. doi: 10.1037/a0020853. [DOI] [PubMed] [Google Scholar]
  21. Kuczmarski R, Ogden C, Guo S, Grummer-Strawn L, Flegal K, Mei Z, Wei R, Curtin L R, Roche AF, Johnson C. 2000 CDC Growth Charts for the United States: methods and development. Vital and Health Statistics. Series 11, Data from the National Health Survey. 2002;246:1–190. [PubMed] [Google Scholar]
  22. Lawman H, Wilson D, Van Horn M, Resnicow K, Kitzman-Ulrich H. The relationship between psychosocial correlates and physical activity in underserved adolescent boys and girls in the ACT Trial. Journal of Physical Activity and Health. 2011;8:253–261. doi: 10.1123/jpah.8.2.253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lawman H G, Wilson D K, Van Horn M L, Resnicow K. Factorial validity and invariance of measures of physical activity motivation and self-efficacy in underserved boys and girls. 2013 Manuscript submitted for publication. [Google Scholar]
  24. MacDonell K, Brogan K, Naar-King S, Ellis D, Marshall S. A pilot study of motivational interviewing targeting weight-related behaviors in overweight or obese African American adolescents. Journal of Adolescent Health. 2012;50:201–203. doi: 10.1016/j.jadohealth.2011.04.018. [DOI] [PubMed] [Google Scholar]
  25. Martin J, McCaughtry N. Using social cognitive theory to predict physical activity in inner-city African American school children. Journal of Sport and Exercise Psychology. 2008;30:378–391. doi: 10.1123/jsep.30.4.378. [DOI] [PubMed] [Google Scholar]
  26. Motl R. Theoretical models for understanding physical activity behavior among children and adolescents—Social cognitive theory and self-determination theory. Journal of Teaching in Physical Education. 2007;26:350–357. [Google Scholar]
  27. Neuhaus J M, Segal M R. Design effect for binary regression model fitted to dependent data. Statistics in Medicine. 1993;12:1259–1268. doi: 10.1002/sim.4780121307. [DOI] [PubMed] [Google Scholar]
  28. Neumark-Sztainer D, Falkner N, Story M, Perry C, Hannan P, Mulert S. Weight-teasing among adolescents: Correlations with weight status and disordered eating behaviors. International Journal of Obesity. 2002;26:123–131. doi: 10.1038/sj.ijo.0801853. [DOI] [PubMed] [Google Scholar]
  29. Ogden C L, Carroll M D, Kit B K, Flegal K M. Prevalence of obesity and trends in body mass index among US children and adolescents, 1999-2010. JAMA. 2012;307:483–490. doi: 10.1001/jama.2012.40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Power T G, Ullrich-French S C, Steele M M, Daratha K B, Bindler R C. Obesity, cardiovascular fitness, and physically active adolescents' motivations for activity: A self-determination theory approach. Psychology of Sport and Exercise. 2011;12:593–598. [Google Scholar]
  31. Puyau M R, Adolph A L, Vohra F A, Zakeri I, Butte N F. Prediction of activity energy expenditure using accelerometers in children. Medicine And Science In Sports And Exercise. 2004;36:1625–1631. [PubMed] [Google Scholar]
  32. Ryan R, Deci E. Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist. 2000;55:68–78. doi: 10.1037//0003-066x.55.1.68. [DOI] [PubMed] [Google Scholar]
  33. Ryan R M, Connell J P. Perceived locus of causality and internalization: Examining reasons for acting in two domains. Journal of Personality and Social Psychology. 1989;57:749–761. doi: 10.1037//0022-3514.57.5.749. [DOI] [PubMed] [Google Scholar]
  34. Sallis J, Grossman R, Pinski R, Patterson T, Nader P. The development of scales to measure social support for diet and exercise behaviors. Preventive Medicine. 1987;16:825–836. doi: 10.1016/0091-7435(87)90022-3. [DOI] [PubMed] [Google Scholar]
  35. Springer A, Kelder S, Hoelscher D. Social support, physical activity and sedentary behavior among 6th-grade girls: A cross-sectional study. International Journal of Behavioral Nutrition and Physical Activity. 2006;3:8. doi: 10.1186/1479-5868-3-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. St. George S M, Wilson D K. A qualitative study for understanding family and peer influences on obesity-related health behaviors in low-income African-American adolescents. Childhood Obesity. 2012;8:466–476. doi: 10.1089/chi.2012.0067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Storch E, Milsom V, DeBraganza N, Lewin A, Geffken G, Silverstein J. Peer victimization, psychosocial adjustment, and physical activity in overweight and at-risk-for-overweight youth. Journal of Pediatric Psychology. 2006;32:80–89. doi: 10.1093/jpepsy/jsj113. [DOI] [PubMed] [Google Scholar]
  38. Teixeira P J, Carraça E V, Markland D, Silva M N, Ryan R M. Exercise, physical activity, and self-determination theory: A systematic review. International Journal of Behavioral Nutrition and Physical Activity. 2012;9:78. doi: 10.1186/1479-5868-9-78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Thomas A S, Greene L F, Ard J D, Oster R A, Darnell B E, Gower B A. Physical activity may facilitate diabetes prevention in adolescents. Diabetes Care. 2009;32:9–13. doi: 10.2337/dc08-0780. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Troiano R, Berrigan D, Dodd K, Masse L, Tilert T, McDowell M. Physical activity in the United States measured by accelerometer. Medicine & Science in Sports & Exercise. 2008;40:181–188. doi: 10.1249/mss.0b013e31815a51b3. [DOI] [PubMed] [Google Scholar]
  41. Trost S, Kerr L, Ward D, Pate R. Physical activity and determinants of physical activity in obese and non-obese children. International Journal of Obesity. 2001;25:822–829. doi: 10.1038/sj.ijo.0801621. [DOI] [PubMed] [Google Scholar]
  42. Trost S G, Sallis J F, Pate R R, Freedson P S, Taylor W C, Dowda M. Evaluating a model of parental influence on youth physical activity. American Journal of Preventive Medicine. 2003;25:277–282. doi: 10.1016/s0749-3797(03)00217-4. [DOI] [PubMed] [Google Scholar]
  43. Ward D, Dowda M, Trost S, Felton G, Dishman R, Pate R. Physical activity correlates in adolescent girls who differ by weight status. Obesity. 2006;14:97–105. doi: 10.1038/oby.2006.12. [DOI] [PubMed] [Google Scholar]
  44. Welk G, Schaben J, Morrow J. Reliability of accelerometry-based activity monitors: A generalizability study. Medicine & Science in Sports & Exercise. 2004;36:1637–1645. [PubMed] [Google Scholar]
  45. Wen C P, Wai J P M, Tsai M K, Yang Y C, Cheng T Y D, Lee M C, Chan H T, Tsao C K, Tsai S P, Wu X. Minimum amount of physical activity for reduced mortality and extended life expectancy: A prospective cohort study. The Lancet. 2011;378:1244–1253. doi: 10.1016/S0140-6736(11)60749-6. [DOI] [PubMed] [Google Scholar]
  46. Wilson D K, Evans A E, Williams J, Mixon G, Sirard J R, Pate R. A preliminary test of a student-centered intervention on increasing physical activity in underserved adolescents. Annals of Behavioral Medicine. 2005;30:119–124. doi: 10.1207/s15324796abm3002_4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Wilson D K, Friend R, Teasley N, Green S, Reaves I L, Sica D A. Motivational versus social cognitive interventions for promoting fruit and vegetable intake and physical activity in African American adolescents. Annals of Behavioral Medicine. 2002;24:310–319. doi: 10.1207/S15324796ABM2404_07. [DOI] [PubMed] [Google Scholar]
  48. Wilson D K, Kitzman-Ulrich H, Williams J E, Saunders R, Griffin S, Pate R, Van Horn M L, Evans A, Hutto B, Addy C L, Mixon G, Sisson S B. An overview of “The Active by Choice Today” (ACT) trial for increasing physical activity. Contemporary Clinical Trials. 2008;29:21–31. doi: 10.1016/j.cct.2007.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Wilson D K, Van Horn M L, Kitzman-Ulrich H, Saunders R, Pate R, Lawman H G, Hutto B, Griffin S, Zarrett N, Addy C L, Mansand L, Mixon G, Brown P V. Results of the “Active by Choice Today”(ACT) randomized trial for increasing physical activity in low-income and minority adolescents. Health Psychology. 2011;30:463–471. doi: 10.1037/a0023390. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Wilson P M, Mack D E, Grattan K P. Understanding motivation for exercise: A self-determination theory perspective. Canadian Psychology. 2008;49:250–256. [Google Scholar]
  51. Wright M, Wilson D, Griffin S, Evans A. A qualitative study of parental modeling and social support for physical activity in underserved adolescents. Health Education Research. 2010;25:224–232. doi: 10.1093/her/cyn043. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Journal of Pediatric Psychology are provided here courtesy of Oxford University Press

RESOURCES