The relationship between transformational teaching and adolescent physical activity: The mediating roles of personal and relational efficacy beliefs

Abstract

The purpose of this study was to examine the extent to which transformational teaching, exhibited by secondary school physical education teachers, predicts within-class physical activity and leisure-time physical activity among adolescents. The study used a prospective observational design and involved data collected from 874 Grade 10 adolescents (M_age = 15.41, (SD) = .61). Through use of structural equation modeling, the results revealed that adolescents’ perceptions of transformational teaching were positively related to within-class physical activity and leisure-time physical activity, and these effects were mediated by adolescents’ estimation of their teacher’s confidence in their abilities (i.e. relation-inferred self-efficacy) and self-efficacy beliefs.

Keywords

adolescence physical activity relational-efficacy beliefs social cognitions structural equation modeling transformational leadership theory

Physical activity among adolescents is associated with improvements in physical and mental health (Hallal et al., 2006). Despite these health benefits, the majority of adolescents in the Western world do not meet current international guidelines for physical activity (Colley et al., 2011; Eaton et al., 2012). As a result, there have been growing calls to understand how to best intervene at the population level, to encourage youth to become and stay active throughout the life span. In light of the considerable amount of time that adolescents spend within school settings, schools have been identified as a critical social context to support engagement in such health-enhancing behaviors (World Health Organization (WHO), 2004), with physical education teachers in particular being well placed to shape adaptive attitudes and motives toward physical activity, and give youth the confidence to adopt and sustain active lifestyles (Bailey et al., 2009).

A useful framework that has been applied to understand the potential mechanisms through which physical education teachers might shape the adoption and maintenance of active lifestyles among adolescents corresponds to transformational leadership theory (Bass and Riggio, 2006; Beauchamp and Morton, 2011). Transformational leadership involves actions that transcend one’s own self-interests with the purpose of empowering and inspiring others to maximize their potential (Bass and Riggio, 2006), and is conceptualized as including four interrelated behavioral dimensions. These include idealized influence, inspirational motivation, intellectual stimulation, and individualized consideration (collectively referred to as the four Is). Idealized influence is characterized by displays of integrity, acting as an appropriate role model, and interacting with others by embodying one’s personally held beliefs (i.e. practicing what we preach). Inspirational motivation involves articulating a compelling vision of the future and displaying enthusiasm and optimism about what others can accomplish. Intellectual stimulation occurs by encouraging others to question current beliefs, addressing challenges from a diverse range of perspectives, and fostering independent thought. Finally, individualized consideration involves recognizing and responding to the individual needs and abilities of others and displaying care, concern, and compassion.

Transformational leadership theory has been applied widely across a range of contexts to understand how those in positions of influence (e.g. business leaders, military commanders, sports coaches, and school teachers) have the capacity to shape the motivational and behavioral responses of others (Barling et al., 2010; Wang et al., 2011). As a basis for supporting the application of transformational leadership theory to education contexts, it has been suggested that both leadership and teaching are concerned with synonymous behavioral processes designed to support the achievement of some set of goals or objectives by another person (Beauchamp and Morton, 2011). In the specific context of school-based physical education, a growing body of evidence suggests that displays of transformational teaching are related to a range of adaptive responses among students. These include improvements in self-determined motivation, psychological need satisfaction, satisfaction with the teacher, enjoyment of physical education, engagement in class, as well as stronger intentions to engage in leisure-time physical activity (LTPA) outside of the school setting (Beauchamp et al., 2010, 2011; Morton et al., 2010; Wilson et al., 2012).

The present study

One of the core theoretical tenets of transformational leadership theory (Bass and Riggio, 2006) is that transformational leaders exert influence over others by giving them the confidence to exceed minimally expected standards. In an early study on the topic, Shamir et al. (1993) suggested that self-efficacy represents a primary motivational mechanism through which transformational leaders influence others. Self-efficacy is defined as a person’s confidence in his or her own capabilities to organize and execute salient task-specific behaviors (Bandura, 1997), and has consistently been implicated as an important cognitive mechanism in the sustained involvement of physical activity behaviors among adolescents (Kohl and Hobbs, 1998). In light of the potential predictive effects of transformational teaching behaviors (by physical education teachers) in shaping elevated self-efficacy beliefs among students, as well as the extant literature linking self-efficacy to physical activity involvement, the present study was concerned with understanding (a) the extent to which transformational teaching, as displayed by physical education teachers, is related to the physical activity behaviors of adolescents, and (b) the extent to which adolescent self-efficacy beliefs might mediate the relationships between transformational teaching and adolescent physical activity.

In addition to testing the relationships between transformational teaching, adolescent self-efficacy, and physical activity, we were also interested in examining how displays of transformational teaching might be able to foster elevated self-efficacy among adolescents. That is, what might be some of the psychological processes among adolescents that result in them feeling more confident in their own capabilities when physical education teachers interact with them through displays of transformational teaching? In order to answer this question, we drew from Lent and Lopez’ (2002) tripartite model of relational efficacy. As an extension of the core tenets of self-efficacy theory (Bandura, 1997), Lent and Lopez suggested that when people perform tasks within interpersonal settings, such as those that exist within therapist–client, coach–athlete, and teacher–student settings, they develop two “types” of relational efficacy beliefs that act as relationship-specific sources of self-efficacy. Specifically, these two efficacy beliefs correspond to other efficacy and relation-inferred self-efficacy (RISE). Other efficacy is conceptualized as “person A’s beliefs about person B’s capabilities” (Lent and Lopez, 2002: 264), which in educational contexts might include a student’s confidence in her teacher’s capabilities. RISE, on the contrary, corresponds to “person A’s appraisals of how person B views person A’s abilities” (Lent and Lopez, 2002: 268). In education settings, this might involve a student’s estimation of whether he or she thinks the teacher is confident in the student’s capabilities.

Recent research by Jackson et al. (2013) provides some support for the contention that both other-efficacy and RISE beliefs held by students in relation to their respective physical education teacher are related to elevated perceptions of self-efficacy. Interestingly, Jackson et al. (2013) also found that those self-efficacy beliefs were also directly and indirectly related to the average amount of time spent each week engaged in exercise by students. Drawing from the tenets of both transformational leadership theory (Bass and Riggio, 2006; Shamir et al., 1993) and Lent and Lopez’ (2002) tripartite model, we hypothesized that when physical education teachers display transformational behaviors in their interactions with students, this would be related to elevated self-efficacy beliefs among students with regard to their within-class capabilities, but that their other-efficacy and RISE beliefs would mediate those relationships (i.e. act as explanatory mechanisms). That is, when teachers make use of the four Is, then students would have greater confidence in their teacher’s capabilities (higher other efficacy) and would be more inclined to think that their teacher believes in them (higher RISE) and that these two forms of relational efficacy would be related to higher within-class self-efficacy beliefs.

In light of an extensive body of evidence that self-efficacy beliefs are related to effortful behavior within physical activity contexts (Kohl and Hobbs, 1998), we further hypothesized that self-efficacy beliefs held by students, with regard to in-class activities, would be related to greater within-class physical activity (WCPA) behavior. Furthermore, in light of recent trans-contextual evidence that elevated self-efficacy beliefs among students within physical education classes are related to greater engagement in leisure-time exercise behavior (Jackson et al., 2013), we also hypothesized that students’ self-efficacy beliefs would be related to greater engagement in LTPA.

When taken together, we hypothesized that physical education teachers’ use of transformational behaviors (i.e. four Is) would be related to greater WCPA and LTPA behavior among students, and that these effects would be mediated by students’ other-efficacy and RISE beliefs, and thereafter adolescent self-efficacy. In order to adequately test each of the abovementioned mediational pathways, both indirect and direct effects were simultaneously modeled. Specifically, if full mediation is supported, the relations between the predictor and the criterion variable in question (i.e. direct effect) would no longer be significant once the effects of the mediator are controlled (cf. Baron and Kenny, 1986). The full path model tested in this study (see Figure 1) was examined through the use of a prospective observational design.

Figure 1.

Path diagram of the hypothesized relationships between transformational teaching, adolescents’ efficacy beliefs, and physical activity behaviors.

Method

Participants

In total, 874 adolescents (N_males = 461; N_females = 413; M_age = 15.41 years, SD = .61 years) from 39 Grade 10 classes across eight schools in the Lower Mainland of British Columbia volunteered to take part in this study. Consistent with procedures used by Statistics Canada in their 2006 Census, participants were asked to identify all ethnic and/or cultural groups with which they self-identified (i.e. students could identify with more than one ethnic group). The largest represented ethnic groups were Canadian (57.2%), Chinese (23.7%), Korean (11.1%), and Indian (5.6%), with 17 other ethnic groups identified with a frequency of less than 5%. The ethnic composition of adolescents sampled in this study was representative of the racial composition of this area of Canada.

Measures

Transformational teaching

Student perceptions of transformational teaching behavior were assessed using the Transformational Teaching Questionnaire (TTQ; Beauchamp et al., 2010). Using this 16-item measure, items are prefixed by “The physical education teacher I am rating …,” with responses recorded on a 5-point rating scale. Anchors include “not at all” (0), “once in a while” (1), “sometimes” (2), “fairly often” (3), and “frequently” (4). The questionnaire consists of four subscales designed to measure the four transformational teaching dimensions of idealized influence (4 items; for example, “Treats me in ways that build my respect”), inspirational motivation (4 items; for example, “Motivates me to try my hardest”), intellectual stimulation (4 items; for example, “Provides me with tasks and challenges that get me to think in different ways”), and individualized consideration (4 items; for example, “Recognizes the needs and abilities of each student in the class”). Results of multilevel confirmatory factor analysis (CFA) provide support for the reliability and factorial validity of measures derived from the TTQ (Beauchamp et al., 2010), including the operationalization of transformational teaching as a higher order measure (i.e. four dimensions contributing to a higher order latent variable). In this study, the higher order (overall) measure of transformational teaching was used with ratings of transformational teaching potentially ranging from 0 to 16. The internal consistency of the transformational teaching measure used in this study was acceptable, with an ordinal coefficient alpha (Zumbo et al., 2007) of .961.

Tripartite efficacy beliefs

Three 9-item instruments that were developed for the specific context of physical education were used to examine students’ self-efficacy, other-efficacy and RISE beliefs (Jackson et al., 2012). In light of measurement recommendations by Myers et al. (2008), as well as those operationalized by Jackson et al. (2012), responses for each of the measures were provided on a 5-point scale with anchors including “no confidence at all” (1), “low confidence” (2), “moderate confidence” (3), “high confidence” (4), and “complete confidence” (5). For self-efficacy, students were asked to “Please honestly rate your confidence in your ability to …,” with exemplar items including “Try your hardest in every PE class,” and “Always listen carefully to your PE teacher’s instructions.” For other efficacy, items were prefixed by “Please honestly rate your confidence in your PE teacher’s ability to …,” with exemplar items including “Motivate you even during hard or unfamiliar activities,” and “Treat all students fairly and equally regardless of how good they are at PE.” Finally, to assess students’ RISE beliefs, students were instructed to “Estimate how confident your PE teacher is in your ability to …,” with exemplar items including, “Perform all the skills you are taught in PE,” and “Try your hardest in every PE class.” Consistent with Jackson et al. (2012), and to emphasize the conceptual separation between self-efficacy and RISE perceptions, students were provided with the following statement prior to the RISE items statement, “so, we’re not focusing on how confident you are; we’re focusing on whether you think your PE teacher is confident in you or not.” The three tripartite measures used in this study have been found to demonstrate adequate reliability, factorial validity, and measurement invariance by gender (Jackson et al., 2012), and in the present study, these measures were found to be internally consistent (self-efficacy ordinal α = .861, other-efficacy ordinal α = .890, and RISE ordinal α = .897).

LTPA

The 2-item WHO Health Behavior in School Children (WHO HBSC) Questionnaire was used to assess students’ LTPA behavior (Booth et al., 2001). Following the prefix “Outside school hours …” adolescents were asked “How often do you usually exercise in your free time, so much that you get out of breath or sweat,” to assess frequency of LTPA on a 1 (once a month or less) to 5 (every day) scale. Using the same prefix, students are then asked “How many hours do you usually exercise in your free time, so much that you get out of breath or sweat” with responses anchored by 1 (none) to 6 (about 7 hours per week). The WHO HBSC LTPA measure has been reported to demonstrate good test–retest reliability and internal consistency, as well as predictive validity in relation to measures derived from the Multistage Fitness Test (Booth et al., 2001) among Grade 8 and 10 students. In the current study, the scores were summed together to provide a total LTPA score that could range between 2 and 11. In this study, this composite measure was found to demonstrate sound internal consistency with a Spearman–Brown coefficient ρ (Eisinga et al., 2012) of .855.

WCPA

WCPA was assessed using the Physical Activity Questionnaire for Adolescents (PAQ-A) developed by Kowalski et al. (1997). The PAQ-A is a self-administered 7-day recall questionnaire that assesses moderate-to-vigorous physical activity during the school year. Measures derived from this instrument have been found to demonstrate sound reliability as well as convergent validity in relation to objective measures of physical activity (Janz et al., 2008). This instrument includes items that assess physical activity behavior during physical education classes, right after school, during evenings, and on the previous weekend. In this study, WCPA was assessed with the 1 item from the PAQ-A, anchored on a 5-point scale, that directly relates to WCPA, namely, “in the last 7-days, during your physical education classes, how often were you very active (playing hard, running, jumping, and throwing)?”

Procedure

Ethical approval was initially obtained from the Behavioral Research Ethics Board at The University of British Columbia, followed by school board approval. After schools had chosen to take part in the study, parents and adolescents were provided with an information letter 3 weeks prior to data collection. Passive consent was obtained from parents, who were provided the opportunity to opt their child out of the study, and children provided active informed consent.

Time 1 (T1) data collection occurred in April 2011, at which point students completed measures of transformational teaching, as well as other-efficacy and RISE beliefs. Time 2 (T2) data collection occurred in June 2011, at which point students completed measures of self-efficacy beliefs and physical activity behavior.

Data analysis

Of the 874 students that completed questionnaires at T1, 718 completed questionnaires at T2 (82.2% retention over time). A missing value analysis was conducted using SPSS (Version 13.0), which indicated that the missing responses at T2 were missing completely at random (MCAR); Little’s (1988) chi-square test was nonsignificant, χ²(27) = 20.73, p = .799. Drawing from procedures outlined by Preacher et al. (2010), we tested the a priori conceptual model through use of structural equation modeling (SEM). Specifically, a random-intercept-only model was specified using Mplus 6.11 (Muthén and Muthén, 1998–2011) to examine direct and indirect effects specified within Figure 1; this model controlled for the nonindependence of observations based on students nested within classes (Asparouhov and Muthén, 2006). A weighted least squares means and variance-adjusted (WLSMV) method of estimation was utilized. This approach incorporates a full information maximum likelihood (FIML) method of estimation at the initial stage of the estimation. The WLSMV method of estimation has been found to appropriately estimate missing data based on assumptions of missing data being missing at random (MAR) (Asparouhov and Muthén, 2010). It should be noted that the SEM approach presented by Preacher et al. (2010) does not require the inclusion of conventional measurement models (e.g. CFA), and instead the observed variables are decomposed into unobserved components, which are considered latent variables.

In the model (Figure 1), structural paths are denoted in accordance with Baron and Kenny’s (1986) conceptual approach to describing mediation. Specifically, paths denoted by a correspond to the relationship between the predictor (transformational teaching) and the mediator variable of interest (e.g. a₁ for other efficacy); paths denoted by b correspond to the effect of the mediator variable on the criterion variable of interest (e.g. b₇ for LTPA); and paths denoted by c′ correspond to the direct effect of the predictor on the outcome variable after controlling for the mediator(s) (e.g. ${c^{'}}_{1}$ for transformational teaching to LTPA). In such instances, the indirect effect is calculated using a × b. In total, nine indirect effects were tested, along with a test of the total indirect effect (Hayes, 2009). It should be noted that while some of these paths tested the effects of a mediator on one of the two dependent variables (e.g. a₁ × b₄), others examined the effects on the dependent variable of interest via a second mediator (e.g. a₂ × b₂ × b₃). Preacher et al. (2010) indicate that no centering is required when this SEM approach is applied to raw data, and thus no centering approach was utilized with the data prior to conducting the analysis. In the structural model, student sex was tested as a covariate. We assessed model fit using χ² goodness-of-fit index, root mean square error of approximation (RMSEA), comparative fit index (CFI), Tucker–Lewis index (TLI), and root mean square residual (RMSR_within).

Results

Descriptive statistics are reported in Table 1. The intraclass correlations (ICCs) for the study variables suggested that transformational teaching (ICC = .23), other efficacy (ICC = .18), and WCPA (ICC = .14) displayed some degree of “groupness”; however, the remaining variables displayed very little group-level variability (ICCs < .06). The fit indices for our a priori hypothesized model were: χ² (17) = 52.307, p < .001, RMSEA = .049, CFI = .963, TLI = .922, and RMSR_within = .018.

Table 1.

Descriptive statistics for the study variables.

Variable	M	SD	Skewness	Kurtosis	ICC
Transformational teaching	10.352	3.498	−0.548	−0.256	0.230
Other efficacy	3.683	0.866	−0.590	−0.066	0.180
RISE	3.812	0.792	−0.696	0.336	0.056
Self-efficacy	3.926	0.708	−0.504	0.060	0.047
WCPA	3.830	0.930	−0.623	0.262	0.141
LTPA	7.30	2.211	−0.512	−0.135	0.026

SD: standard deviation; ICC: intraclass correlation; RISE: relation-inferred self-efficacy; WCPA: within-class physical activity; LTPA: leisure-time physical activity.

Consistent with an extensive body of epidemiological evidence that has examined gender differences in social cognitions and physical activity behavior among adolescents, boys were found to display greater levels of self-efficacy with regard to their within-class activities (b = −.181, p < .001) as well as higher levels of physical activity in terms of both LTPA (b = −.485, p = .012) and WCPA (b = −.330, p = .003) (see Table 2, Figure 2). Boys also believed that their teachers had greater confidence in them (high RISE) than girls (b = −.314, p < .001). In terms of the main analyses (see Table 2, Figure 2), and after controlling for student sex as a covariate, the results revealed that displays of transformational teaching were able to predict significant variance in WCPA behavior, and that this relationship was mediated by students’ RISE beliefs and self-efficacy, a₂ × b₂ × b₃ (b = 0.049, p < .001), with 20% of the variance in WCPA explained by the model (R² = .200). Notably, after this multiple mediation path was accounted for in the structural model, the direct effect between transformational teaching and WCPA was found to be nonsignificant, thereby providing evidence of full mediation. The results revealed that displays of transformational teaching were also able to explain significant variance in adolescents’ LTPA behavior, with two multiple mediator pathways contributing to the model. Specifically, transformational teaching was related to LTPA via students’ RISE beliefs, a₂ × b₈ (b = .063, p = .001), as well as via RISE beliefs and self-efficacy, a₂ × b₂ × b₆ (b = .088, p < .001). As with WCPA, once these mediating paths were controlled in the model, the direct effects of transformational teaching in relation to LTPA were nonsignificant, which provides evidence of full mediation. In total, the model explained 25.3% of the variance in adolescents’ LTPA behavior (R² = .253). The total indirect effect was also significant (b = .185, p < .001).

Table 2.

Structural equation model of the effects of transformational teaching on student physical activity.

Variables	Estimate	SE	p	95% CI
Direct effects (within level)
Model A (outcome: OE)
Predictor: TT (a₁)	0.196	0.004	<.001	0.188 to 0.204
Covariate: sex	−0.046	0.056	.417	−0.156 to 0.065
Model B (outcome: RISE)
Predictor: TT (a₂)	0.128	0.006	<.001	0.116 to 0.140
Covariate: sex	−0.314	0.060	<.001	−0.431 to −0.197
Model C (outcome: SE)
Mediator: OE (b₁)	0.011	0.028	.683	−0.043 to 0.066
Mediator: RISE (b₂)	0.545	0.022	<.001	0.502 to 0.589
Predictor: TT (a₃)	0.012	0.008	.154	−0.004 to 0.027
Covariate: sex	−0.181	0.028	<.001	−0.236 to −0.125
Model D (outcome: WCPA)
Mediator 1: SE (b₃)	0.696	0.073	<.001	0.554 to 0.838
Mediator 2: OE (b₄)	0.059	0.129	.646	−0.193 to 0.311
Mediator 3: RISE (b₅)	−0.083	0.079	.294	−0.238 to 0.072
Predictor: TT (c′)	−0.019	0.028	.515	−0.074 to 0.037
Covariate: sex	−0.330	0.113	.003	−0.550 to −0.109
Model E (outcome: LTPA)
Mediator 1: SE (b₆)	1.256	0.156	<.001	0.950 to 1.561
Mediator 2: OE (b₇)	−0.214	0.169	.204	−0.545 to 0.116
Mediator 3: RISE (b₈)	0.492	0.158	.002	0.184 to 0.801
Predictor: TT (c′)	−0.109	0.056	.050	−0.218 to 0
Covariate: sex	−0.485	0.192	.012	−0.862 to −0.109
Indirect effects (within level)
Ind1 = a₁ × b₁ × b₃	0.002	0.004	.683	−0.006 to 0.009
Ind2 = a₂ × b₂ × b₃	0.049	0.007	<.001	0.035 to 0.062
Ind3 = a₁ × b₄	0.012	0.025	.646	−0.038 to 0.061
Ind4 = a₂ × b₅	−0.011	0.010	.296	−0.031 to 0.008
Ind5 = a₃ × b₃	0.008	0.006	.163	−0.003 to 0.019
Ind6 = a₁ × b₁ × b₆	0.003	0.007	.685	−0.011 to 0.016
Ind7 = a₂ × b₂ × b₆	0.088	0.014	<.001	0.060 to 0.115
Ind8 = a₁ × b₇	−0.042	0.033	.204	−0.107 to 0.023
Ind9 = a₂ × b₈	0.063	0.019	.001	0.025 to 0.101
Ind10 = a₃ × b₆	0.014	0.010	.166	−0.006 to 0.035
Total indirect effect	0.185	0.041	<.001	0.104 to 0.266

TT: transformational teaching; SE: self-efficacy; OE: other efficacy; RISE: relation-inferred self-efficacy; WCPA: within-class physical activity; LTPA: leisure-time physical activity; CI: confidence interval.

This is a random-intercept-only multilevel model, and so the results are reported at the within-level (i.e. student-level) only. The indirect effects refer to the mediational pathways as specified in Figure 1. For example, Ind2 corresponds to a₂ *b₂ *b₃ (TT→RISE→SE→WCPA).

Figure 2.

Path diagram of the relationships between transformational teaching, adolescents’ efficacy beliefs, and physical activity behaviors.

Discussion

The overall purpose of this study was to examine the extent to which adolescents’ perceptions of their school physical education teacher’s transformational teaching behaviors are prospectively related to their own physical activity behaviors, as well as the extent to which their personal and relational efficacy beliefs mediate those relationships. Overall, the results revealed that when teachers made use of the four Is (idealized influence, inspirational motivation, intellectual stimulation, and individualized consideration), students reported more effortful behavior during class time, and that this relationship was mediated via students’ RISE and self-efficacy beliefs. Specifically, this multiple mediational pathway suggests that when teachers make use of transformational teaching behaviors, adolescents are more likely to think that their respective teacher believes in them (high RISE), and subsequently display elevated self-efficacy beliefs, and thereafter WCPA behavior. Although previous research from the field of organizational psychology has provided evidence for the effects of transformational leadership in relation to follower self-efficacy beliefs (Pillai and Williams, 2004), this study represents the first to demonstrate that RISE beliefs act as an explanatory mechanism of those effects.

In addition to the effects with regard to within-class behavior, adolescents’ perceptions of their physical education teacher’s behaviors also predicted significant variance in LTPA behavior, which was mediated by adolescents’ RISE and self-efficacy beliefs. Evidence for two multiple mediator pathways was found in the prediction of LTPA. Specifically, by interacting with their students through displays of transformational teaching, physical education teachers were believed by their students to have greater confidence in their (i.e. students’) capabilities. These RISE beliefs were not only associated with students having greater confidence in their own within-class capabilities (i.e. higher self-efficacy), and physical activity behavior outside of the school environment (i.e. transformational teaching→RISE→self-efficacy→LTPA), RISE also had a direct effect in relation to LTPA that was unmediated by self-efficacy. When taken together, these findings provide support for the contention that school physical education teachers have the potential to shape adolescents’ involvement in physical activity outside of school hours (Beauchamp and Morton, 2011). Indeed, the findings suggest that through the use of transformational teaching behaviors, secondary school physical education teachers are able to promote important health-enhancing social cognitions (e.g. RISE and self-efficacy) that can shape (a) students’ in-class activity patterns, as well as (b) choices to engage in effortful LTPA when such actions are under volitional control.

Although both RISE and self-efficacy beliefs were found to be involved in the prediction of both WCPA and LTPA behavior, it is also noteworthy that other efficacy did not emerge as a salient mediator of the relations between transformational teaching and either of the two physical activity outcome measures. That is, although displays of transformational teaching predicted students’ confidence in their teacher’s capabilities, those other-efficacy beliefs did not further predict any significant variance in adolescent physical activity behavior. This finding stands in contrast to previous research by Jackson et al. (2013), which found that other-efficacy beliefs were related to students’ within-class self-efficacy beliefs as well as the average amount of time spent exercising over a 3-week period. In light of these findings, future research appears warranted that examines the extent to which students’ confidence in their teacher’s capabilities (i.e. high other efficacy) is implicated in their own social cognitions related to their own capabilities as well as subsequent engagement behaviors.

Balanced against the potential contributions and insight provided by these findings, limitations to the present study should also be acknowledged. First, it should be noted that the criterion variables operationalized in this study were assessed through use of brief self-report measures. Both the WHO HBSC questionnaire (Booth et al., 2001) and PAQ-A (Kowalski et al., 1997) have been used extensively to assess adolescent physical activity behaviors within surveillance research, and have resulted in consistent displays of reliability and validity evidence. Nevertheless, in light of the potential for such measures to result in overestimation of actual behavioral indices, future research is clearly warranted that makes use of objective measures of physical activity, including accelerometry (Trost et al., 2000) and Global Positioning System (GPS) monitoring (Maddison and Mhurchu, 2009). As a second limitation, although we utilized a prospective design, such approaches still preclude any inferences of causality. Previous research within educational settings with physical education teachers suggests that transformational teaching behaviors can be developed through intervention (Beauchamp et al., 2011). Future research would seem well placed to examine the extent to which changes in transformational teaching (through intervention) might result in changes in social cognitions among teenagers, such as RISE and self-efficacy beliefs, and thereafter adolescent physical activity behavior, both with respect to within-class and leisure-time behaviors.

Footnotes

Funding

The research presented in this paper was supported by a career investigator award from the Michael Smith Foundation for Health Research (Award # CI-SCH-01870(07-1) and a Standard Research Grant from the Social Sciences and Humanities Research Council of Canada (Award # 410-2006-0389) awarded to Mark Beauchamp.

References

Asparouhov

Muthén

(2006) Comparison of estimation methods for complex survey data analysis (Technical Report). Available at: http://pages.gseis.ucla.edu/faculty/muthen/articles/Article_110.pdf (accessed 1 December 2012).

Asparouhov

Muthén

(2010) Weighted least squares estimation with missing data (Technical Report). Available at: http://www.statmodel.com/download/GstrucMissingRevision.pdf (accessed 1 December 2012).

Bailey

Armour

Kirk

. (2009) The educational benefits claimed for physical education and sport school: An academic review. Research Papers in Education 24: 1–27.

Bandura

(1997) Self-Efficacy: The Exercise of Control. New York: W.H. Freeman & Co.

Barling

Christie

Hoption

(2010) Leadership. In: Zedeck

(ed.) Handbook of Industrial and Organizational Psychology. Washington, DC: APA Books.

Baron

Kenny

(1986) The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology 51: 1173–1182.

Bass

Riggio

(2006) Transformational Leadership (2nd edn). Mahwah, NJ: Erlbaum.

Beauchamp

Morton

(2011) Transformational teaching and physical activity engagement among adolescents. Exercise and Sport Sciences Reviews 39: 133–139.

Beauchamp

Barling

Morton

(2011) Transformational Teaching and adolescent self-determined motivation, self-efficacy, and intentions to engage in leisure time physical activity: A randomized controlled pilot trial. Applied Psychology: Health and Well-Being 3: 127–150.

10.

Beauchamp

Barling

. (2010) Development and psychometric properties of the transformational teaching questionnaire. Journal of Health Psychology 15: 1123–1134.

11.

Booth

Okely

Chey

. (2001) The reliability and validity of the physical activity questions in the WHO health behaviour in schoolchildren (HBSC) survey: A population study. British Journal of Sports Medicine 25: 263–267.

12.

Colley

Garriguet

Janssen

. (2011) Physical activity of Canadian children and youth: Accelerometer results from the 2007 to 2009 Canadian Health Measures Survey. Statistics Canada, Health Reports, vol. 22, pp. 1–8. Available at: http://www.statcan.gc.ca/pub/82-003-x/2011001/article/11397-eng.pdf

13.

Eaton

Kann

Kinchen

. (2012) Youth risk behavior surveillance—United States, 2011. Morbidity and Mortality Weekly Report, Surveillance Summaries 61(SS04): 1–162. Available at: http://www.cdc.gov/mmwr/pdf/ss/ss6104.pdf (accessed 5 December 2012).

14.

Eisinga

Te Grotenhuis

Pelzer

(2012) The reliability of a two-item scale: Pearson, Cronbach or Spearman-Brown? International Journal of Public Health. Epub ahead of print. DOI: 10.1007/s00038-012-0416-3.

15.

Hallal

Victora

Azevedo

. (2006) Adolescent physical activity and health: A systematic review. Sports Medicine 36: 1019–1030.

16.

Hayes

(2009) Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs 76: 408–420.

17.

Jackson

Whipp

Beauchamp

(2013) The tripartite efficacy framework in high school physical education: Trans-contextual generality and direct and indirect prospective relations with leisure-time exercise. Sport, Exercise, and Performance Psychology 2: 1–14.

18.

Jackson

Whipp

Chua

KLP

. (2012) Assessment of tripartite efficacy beliefs within school-based physical education: Instrument development and reliability and validity evidence. Psychology of Sport and Exercise 13: 108–117.

19.

Janz

Lutuchy

Wenthe

. (2008) Measuring activity in children and adolescents using self-report: PAQ-C and PAQ-A. Medicine and Science in Sports and Exercise 40: 767–772.

20.

Kohl

Hobbs

(1998) Development of physical activity behaviors among children and adolescents. Pediatrics 101: 549–554.

21.

Kowalski

Crocker

PRE

Kowalski

(1997) Convergent validity of the Physical Activity Questionnaire for Adolescents. Pediatric Exercise Science 9: 342–352.

22.

Little

(1988) A test of missing completely at random for multivariate data with missing values. Journal of the American Statistical Association, 83: 1198–1202

23.

Lent

Lopez

(2002) Cognitive ties that bind: A tripartite view of efficacy beliefs in growth-promoting relationships. Journal of Social and Clinical Psychology 21: 256–286.

24.

Maddison

Mhurchu

(2009) Global positioning system: A new opportunity in physical activity measurement. International Journal of Behavioral Nutrition and Physical Activity 6: 73.

25.

Morton

Keith

Beauchamp

(2010) Transformational teaching and physical activity: A new paradigm for adolescent health promotion? Journal of Health Psychology 15: 248–257.

26.

Muthén

(1998–2011) Mplus User’s Guide (6th edn). Los Angeles, CA: Muthén & Muthén.

27.

Myers

Feltz

Wolfe

(2008) A confirmatory study of rating scale category effectiveness for the coaching efficacy scale. Research Quarterly for Exercise and Sport 79: 300–311.

28.

Pillai

Williams

(2004) Transformational leadership, self-efficacy, group cohesiveness, commitment, and performance. Journal of Organizational Change Management 17: 144–159.

29.

Preacher

Zyphur

Zhang

(2010) A general multilevel SEM framework for assessing multilevel mediation. Psychological Methods 15: 209–233.

30.

Shamir

House

Arthur

(1993) The motivational effects of charismatic leadership: A self-concept based theory. Organization Science 4: 577–594.

31.

Statistics Canada (2006) Ethnocultural portrait of Canada highlight tables, 2006 census. Available at: http://www12.statcan.ca/census-recensement/2006/dp-pd/hlt/97-562/index.cfm?Lang=E (accessed 16 January 2013).

32.

Trost

Pate

Freedson

. (2000) Using objective physical activity measures with youth: How many days of monitoring are needed? Medicine and Science in Sports and Exercise 32: 426–431.

33.

Wang

Courtright

. (2011) Transformational leadership and performance across criteria and levels: A meta-analytic review of 25 years of research. Group Organization Management 36: 223–270.

34.

Wilson

Liu

Keith

. (2012) Transformational teaching and child psychological needs satisfaction, motivation, and engagement in elementary school physical education. Sport, Exercise, and Performance Psychology 1: 215–230.

35.

World Health Organization (WHO) (2004) Global strategy on diet, physical activity and health. Available at: http://www.who.int/dietphysicalactivity/strategy/eb11344/strategy_english_web.pdf (accessed 16 January 2013).

36.

Zumbo

Gadermann

Zeiser

(2007) Ordinal versions of coefficients alpha and theta for Likert rating scales. Journal of Modern Applied Statistical Methods 6: 21–29. Available at: http://educ.ubc.ca/faculty/zumbo/papers/ordinal_alpha_reprint.pdf (accessed 16 January 2013).