teachers and ECTs to deliver high quality physical activity programs. As specified in the Student Health .... Caucasian (81%), and master's degree holders (64%).
Journal of Teaching in Physical Education Journal of Teaching in Physical Education, 2013, 32, 419-440 © 2013 Human Kinetics, Inc.
Endorsed by the Curriculum and Instruction Academy of the NASPE and the AIESEP www.JTPE-Journal.com ARTICLE
Elementary Classroom Teachers’ Adoption of Physical Activity Promotion in the Context of a Statewide Policy: An Innovation Diffusion and Socio-Ecologic Perspective Collin Andrew Webster1, Peter Caputi2, Melanie Perreault1, Rob Doan1, Panayiotis Doutis1, and Robert Glenn Weaver1 1University
of South Carolina; 2University of Wollongong
Physical activity promotion in the academic classroom (PAPAC) is an effective means for increasing children’s school-based physical activity. In the context of a South Carolina policy requiring elementary schools to provide children with 90 min of physical activity beyond physical education every week, the purpose of this study was to test a theoretical model of elementary classroom teachers’ (ECT) PAPAC adoption drawing from Rogers’ (1995) diffusion of innovations theory and a social ecological perspective. ECTs (N = 201) were assessed on their policy awareness, perceived school support for PAPAC, perceived attributes of PAPAC, domain-specific innovativeness, and self-reported PAPAC. Partial least squares analysis supported most of the hypothesized relationships. Policy awareness predicted perceived school support, which in turn predicted perceived attributes and domain-specific innovativeness. Perceived compatibility, simplicity, and observability, and domain-specific innovativeness predicted self-reported PAPAC. This study identifies variables that should be considered in policy-driven efforts to promote PAPAC adoption. Keywords: primary teachers, nonspecialists, classroom-based physical activity, social ecological model, diffusion of innovations theory
School-based physical activity promotion is one of the key recommended strategies for increasing children’s daily physical activity (US Department of Health and Human Services [USDHHS], 2001). The National Association for Sport and Physical Education (NASPE, 2008) currently holds that physical education should play a central role in school-based physical activity promotion, but also that children should be provided with additional opportunities to be active throughout the school Webster, Perreault, Doan, Doutis, and Weaver are with Physical Education and Athletic Training Dept., University of South Carolina, Columbia, SC. Caputi is with the Psychology Dept., University of Wollongong, Wollongong, New South Wales, Australia. 419
420 Webster et al.
day. Consistent with these recommendations, the South Carolina Student Health and Fitness Act of 2005 (www.scstatehouse.gov) requires all state funded elementary schools to provide children with 60 min of specialist-taught physical education and 90 min of additional physical activity every week. As part of this policy, considerations for implementing the additional physical activity requirement include the involvement of generalist elementary classroom teachers (ECT). Involving ECTs in school wide physical activity promotion is congruent with guidelines in the education sector of the US National Physical Activity Plan (www.physicalactivityplan. org), which recommend requiring continuing education for both physical education teachers and ECTs to deliver high quality physical activity programs. As specified in the Student Health and Fitness Act, one of the contexts where ECTs are encouraged to promote physical activity is in the academic classroom. Physical activity promotion in the academic classroom (PAPAC) is a common focus in both scholarly and professional literature on school wide physical activity promotion (e.g., Hall, Little, & Heidorn, 2011; Rink, Hall, & Williams, 2010; Webster, 2011; Webster, Erwin, & Parks, in press; Webster, Monsma, & Erwin, 2010). Examples of recommended classroom-based strategies in the Student Health and Fitness Act include providing activity breaks between classroom instruction and integrating physical activity into academic lessons. Studies have demonstrated that PAPAC is a viable and effective means for increasing children’s physical activity (Bartholomew & Jowers, 2011, Erwin, Beighle, Morgan, & Noland, 2011; Holt, Heelan, & Bartee, 2012; Mahar et al., 2006). These studies showed that teachers using packaged programs (e.g., Texas I-CAN!) were able to significantly increase their students’ in-class physical activity. Moreover, PAPAC is particularly attractive amid the current accountability context in schools for academic achievement, as studies have shown PAPAC can improve students’ classroom performance and learning outcomes (Bartholomew & Jowers, 2011; Erwin, Fedewa, Beighle, & Ahn, 2012; Mahar et al., 2006). Despite its apparent effectiveness as a method to increase children’s daily physical activity, and its facilitative role in currently prioritized academic goals in schools, little is known about the factors that relate to ECTs’ PAPAC. Bartholomew and Jowers (2011) found that ECTs implemented more Texas I-CAN! lessons when they perceived the lessons to be consistent with required curricular content, perceived higher self-efficacy for implementing the lessons, and perceived fewer barriers to implementation (e.g., lack of time). The authors also emphasized links between their findings and key variables from the Theory of Planned Behavior (Ajzen, 1985; e.g., teacher attitudes, perceived behavioral control). More generally, research has shown that the implementation of new programs in schools can depend on numerous variables, such as teachers’ perceptions of a supportive school climate and beliefs about the program (Beets et al., 2008). However, mechanisms that underpin PAPAC in the context of policy implementation are relatively unknown. Investigating these mechanisms is an important step in constructing the knowledge base needed to translate policy to practice, and can help to reveal important variables that should be targeted in PAPAC training and interventions emerging against the backdrop of the National Physical Activity Plan. In the current study, we drew on diffusion of innovations theory (Rogers, 1995) and a social ecological perspective from the health promotion field (McLeroy, Bibeau, Steckler, & Glanz, 1988; Spence & Lee, 2003) to investigate possible predictors of South Carolina ECTs’ adoption of PAPAC.
Adoption of Physical Activity Promotion 421
Diffusion of Innovations Theory Rogers’ (1995) diffusion of innovations theory explains the adoption and diffusion of new practices within a given social system. The theory has found useful application in research from a broad range of fields, including education and public health (Dearing, 2009), and more specifically in school-based physical activity interventions (McKenzie, Sallis, & Rosengard, 2009; Owen, Glanz, Sallis, & Kelder, 2006). Adoption is defined as the process of a potential adopter becoming aware of, and eventually using a given innovation on a regular basis, whereas diffusion is defined as widespread adoption of the innovation among members of the social system (Metzler, Lund, & Gurvitch, 2008). The first step in the adoption process is awareness and knowledge of the innovation, which allows the potential adopter to evaluate the innovation and decide whether to adopt it (Rogers, 1995). In the context of the South Carolina Student Health and Fitness Act, the more ECTs become aware of and knowledgeable about the state policy, the more they should recognize that there is a requirement for 90 additional minutes of physical activity every week beyond physical education, understand that considerations for implementing this requirement include PAPAC, and know which PAPAC strategies are specified in the policy. In the next stage of the adoption process, Rogers (1995) suggests that potential adopters evaluate the merits of the innovation when faced with a decision about adoption. Rogers (1995) contends that potential adopters are more likely to feel persuaded to, ultimately, fully adopt an innovation when they perceive it as having five attributes: relative advantage, compatibility, simplicity, trialability, and observability. Relative advantage is the perception that the innovation has advantages relative to current practices the potential adopter is using. Compatibility is the perception that the innovation is compatible with the potential adopter’s values, past experiences, and needs as they relate to his/her role within the context in question (e.g., academic classroom setting). Simplicity is the perception that the innovation is easy to understand and use. Trialability is the perception that the innovation can be gradually adopted on an experimental basis, rather than needing to be immediately adopted in full. Observability is the perception that the innovation will produce observable results for key members of the social system (e.g., school). Therefore, once ECTs have gained awareness and knowledge of the Student Health and Fitness Act, it would be in line with diffusion of innovations theory to suggest they might then consider the perceived attributes of PAPAC before fully adopting it. For instance, given the pressure placed on schools and teachers for students’ academic achievement, it would be reasonable to assume that many ECTs might evaluate PAPAC in terms of the advantages they perceive it having as a vehicle for generating high test scores. In addition to considering the perceived attributes of an innovation, potential adopters are more likely to adopt the innovation when they, themselves are more innovative (Rogers, 1995). Innovativeness is defined as “the degree to which an individual is earlier in adopting new ideas than the average member of his or her social system” (Rogers & Shoemaker, 1971). While it has not received much investigation in health promotion research, innovativeness has been widely examined in consumerism studies (e.g., Bass, 1969; Craig & Ginter, 1975; Goldsmith & Hofacker, 1991; Hirschman, 1980; Joseph & Vyas, 1984; Rogers & Shoemaker, 1971). Ho and Wu (2011) suggest that, within the consumerism literature, the
422 Webster et al.
perceived attributes of an innovative product and consumer innovativeness are the two key predictors of adoption. Consumerism researchers distinguish between innovativeness as a trait characteristic (i.e., general personality characteristic) and as a domain-specific characteristic (i.e., context-driven characteristic; Citrin, Sprott, Silverman, & Stem, 2000). Research using measures of both general and domain-specific innovativeness has shown domain-specific innovativeness to be a more useful predictor of adoption than general innovativeness, which does not overlap much across domains or contexts (Citrin et al., 2000; Gatignon & Robertson, 1985; Goldsmith & Hofacker, 1991; Hirschman, 1980). According to Citrin et al. (2000), domain-specific innovation “reflects the tendency to learn about and adopt innovations within a specific domain or interest and, therefore, taps a deeper construct of innovativeness more specific to an area of interest” (p. 296). From this perspective, it stands to reason that South Carolina ECTs with more educational innovativeness might be more receptive and proactive when introduced to new educational policies, and therefore be more inclined to adopt physical activity promotion strategies in response to the Student Health and Fitness Act.
Social Ecological Perspective Research using diffusion of innovations theory has shown that it is important to also investigate contextual factors that can influence the adoption of an innovation. Organizational context, such as resource availability and organizational climate, has played a powerful role in the adoption of health promotion programs in schools (Deschesnes, Trudeau, & Kébé, 2009; Lochman, 2003; Ringeisen, Henderson, & Hoagwood, 2003). Lochman (2003) cites research in which perceived school environment and perceived support from school administrators were related to the successful implementation of new programs. One approach that can be used to better account for the role of organizational context in the adoption process is to consider this variable in relation to policy awareness, perceived attributes of PAPAC, and domain-specific innovativeness as part of a socioecologic framework. Bronfenbrenner (1977, 1979) proposed a social ecological perspective to identify interrelated factors that operate at different levels to influence human behavior in a particular environment. In the health promotion field, McLeroy et al.’s (1988) social ecological model, which is an adaptation of Bronfenbrenner’s (1977, 1979) work, has helped to illuminate the integrated nature of multiple levels of influence on school-based physical activity promotion (Langille & Rodgers, 2010). Three of these levels—intrapersonal, institutional, and policy—are particularly relevant to the diffusion of innovations variables and research presented above. Intrapersonal factors include characteristics of the individual whose behavior is being targeted through promotion efforts. Examples of such characteristics include attitudes, selfconcept, beliefs, skills, and perceptions related to the innovation. ECTs’ perceived attributes of PAPAC and domain-specific innovativeness fit logically into this level of McLeroy et al.’s (1988) model. Institutional factors include characteristics of the organization in which the individual works. Such characteristics can include the physical and social environment of the organization. In the context of the current study, a supportive school environment for PAPAC might include the availability of relevant classroom materials, enough classroom space, and support from the school
Adoption of Physical Activity Promotion 423
principal. Public policy factors include local, state, or national policies (McLeroy et al., 1988). ECTs’ awareness of the Student Health and Fitness Act aligns with this level of McLeroy et al.’s (1988) model. Social ecological perspectives assume the nature of a variable’s influence on a given target behavior (e.g., PAPAC) changes from relatively proximal/direct to relatively distal/indirect depending on where the variable is situated in the model (Spence & Lee, 2003). Intrapersonal variables exist within the individual whose behavior is being targeted, so these variables are expected to have the most proximal/direct influence on the behavior. Both institutional and public policy variables surround the individual, but institutional variables are expected to have a more immediate and proximal/direct influence on the individual’s behavior than public policy variables (Spence & Lee, 2003). Consistent with this view, it would seem ECTs’ perceived attributes of PAPAC and domain-specific innovativeness should be directly related to these teachers’ use of PAPAC strategies, whereas school support for PAPAC should be indirectly related to PAPAC through perceived attributes and domain-specific innovativeness. Furthermore, ECTs’ awareness of the Student Health and Fitness Act should be situated at the most distal layer in the model, indirectly influencing the teachers’ use of PAPAC strategies through both the institutional and intrapersonal layers of influence.
Purpose of the Study The purpose of this study was twofold. Given previous research on ECTs’ PAPAC has not investigated the variables of interest in this study, a preliminary aim was to test the psychometric properties of the measures employed in this study and establish their validity and reliability with the current sample. A further aim was to then test a theoretical model of South Carolina ECTs’ adoption of PAPAC, drawing on the diffusions of innovations and socioecologic perspective discussed above. The specific research question driving this study was, “What are the relationships among ECTs’ awareness of the Student Health and Fitness Act, perceived school support for PAPAC, perceived attributes of PAPAC, domain-specific innovativeness, and self-reported PAPAC?” We hypothesized direct and positive relationships between policy awareness and perceived school support, perceived school support and intrapersonal factors, and intrapersonal factors and self-reported PAPAC. We also hypothesized indirect and positive relationships between policy awareness and intrapersonal variables, policy awareness and self-reported PAPAC, and perceived school support and self-reported PAPAC.
Method Participants and Setting Participants (N = 201, mean age = 41.46, SD = 11.35) were ECTs employed in South Carolina schools. These teachers reported having an average of 14.57 (SD = 9.31) years of elementary classroom teaching experience in total and 13.43 (SD = 9.04) years of elementary classroom teaching experience in South Carolina. The majority of the participants were female (92%), White Caucasian (81%), and master’s degree holders (64%).
424 Webster et al.
For participants who reported their grade level, the breakdown was prekindergarten (10%), kindergarten (9.5%), grade 1 (7%), grade 2 (6.5%), grade 3 (18.9%), grade 4 (14.4%), grade 5 (12.9%), and grade 6 (1.5%). In terms of classroom context, 93% of participants reported having no more than 28 students (the South Carolina Health and Fitness Act cites a 28:1 student-teacher ratio as average for public elementary schools in the state). In addition, 66% of participants reported having a full time teaching assistant, 93% reported having a part time teaching assistant, and 78% reported having parent assistants. Based on the data provided, ECTs in this study were employed at 79 schools (out of 638) in South Carolina. The majority of the schools (76%) represented in the study were Title 1 schools. There were 42 school districts (out of 103) represented, spanning a broad geographic region of the state. Of those schools represented in the study, the breakdown for school district locations was 63% rural, 27% urban, and 1% suburban, which is similar to the general demographic profile for school districts in the state.
Research Design and Instrumentation A descriptive/cross-sectional research design was employed in this study. The intent behind using this design was to ascertain a “snapshot” of South Carolina ECTs’ PAPAC at a single time point (Thomas, Nelson, & Silverman, 2011). Although a longitudinal design would allow for investigation of the adoption process over time, the current study was designed only to test proposed relationships among variables according to the previously outlined theoretical framework. Questionnaires were employed in this study, given the aim to assess ECTs’ PAPAC across South Carolina in the context of a statewide policy (Thomas et al., 2011). The questionnaires used for this study are described below. Since the variables investigated in this study have not been assessed or tested in previous research on ECTs’ PAPAC, we adapted several measures from previously established instruments used in other fields or health contexts, where appropriate. In cases where we were unable to identify any existing instruments that were relevant to the assessment of a given variable, we developed our own measures. Policy Awareness. At the outset of the questionnaire packet used in this study,
participants were instructed to read the following statement: “National and state policies identify the academic classroom as one place where children should be provided opportunities to be physically active.” Congruent with this statement, we developed a single item to assess the extent of ECTs’ awareness of the Student Health and Fitness Act. A preamble for the item read: “In 2005, South Carolina passed legislation requiring all elementary schools to provide children with 60 minutes of physical education and 90 additional minutes of physical activity each week. The item then read, “How much did you already know about this policy?” Participants responded on a 4-point scale that used the following response options: “Was not aware of the policy,” “Just a little,” “Knew most of it,” and “Knew all of it.”
Perceived School Support. We developed an 8-item questionnaire for this study to assess ECTs’ perceived school support. Following guidelines from Kreuger and Casey (2000), preliminary focus group interviews were conducted with a
Adoption of Physical Activity Promotion 425
different sample of ECTs (N = 15) from two state funded elementary schools in South Carolina to develop items for the questionnaire. Publicly available school performance data were used to select a high performing and a low performing school in an attempt to interview teachers working in contexts that were more and less likely to have a supportive school environment for PAPAC. Given the limited availability of research on the relationship between school characteristics and ECTs’ perspectives on PAPAC, we felt that school performance might serve to distinguish schools on a range of characteristics that could potentially influence PAPAC. Focus group participants responded to the questions, “What aspects of your school environment, either physical or social, help you to promote physical activity?” and “What aspects do you see as obstacles to physical activity promotion at your school?” Based on the interview responses, we wrote the questionnaire items to target various sources of support, including classroom materials (“I have enough materials to provide opportunities for children to be physically active in my classroom”), the physical environment of the classroom (e.g., “My classroom environment can be easily modified to provide opportunities for my children to be physically active”), the students in the classroom (e.g., “My students make it easy for me to provide them with opportunities to be physically active in my classroom”), the school schedule and curriculum (e.g., “The school schedule allows me to provide children with opportunities to be physically active in my classroom”) and the school administration (“Overall, my school administration poses a barrier to providing opportunities for children to be physically active in my classroom”). The questionnaire uses a 7-point response scale with the anchors 1 = “Strongly disagree” and 7 = “Strongly agree.” Perceived Attributes of PAPAC. We adapted a questionnaire from school-based health promotion research (Pankratz, Hallfors, & Cho, 2002) to assess ECTs’ perceived attributes of PAPAC. Pankratz et al. (2002) examined policy coordinators’ perceived attributes of a school-based drug prevention innovation. The researchers developed a 17-item questionnaire, using a five-point response scale with the anchors 1 = “Strongly disagree” and 5 = “Strongly agree” to assess all five perceived attributes consistent with diffusion of innovations theory (Rogers, 1995). The initial factor loadings resulted in conceptual problems and low internal consistency for the three items designed to assess trialability. Therefore, Pankratz et al. (2002) eliminated these items and retained a three-factor solution they believed was conceptually meaningful. The first factor included items assessing relative advantage and compatibility (8 items), the second factor included items assessing complexity (4 items), and the third factor included items assessing observability (2 items). We made several modifications to the original instrument to align with the purpose of the current study and strengthen the measure. These included recontextualizing Pankratz et al.’s (2002) final 14 items to fit the academic classroom setting, changing the focus of the innovation to PAPAC, drawing on Rogers’ (1995) definitions of relative advantage, compatibility, trialability, and observability to rewrite/develop items for these constructs, and increasing the scale range to seven points to allow for more response variability. Our modifications resulted in an 18-item questionnaire designed to assess all five perceived attributes: relative advantage (4 items), compatibility (3 items),
426 Webster et al.
complexity (4 items), trialability (4 items), and observability (3 items). The modified measure uses the same anchor terminology as the original measure. Example items for each variable are “Providing opportunities for children to be physically active in my classroom will increase the quality of education my students receive” (relative advantage), “Providing opportunities for children to be physically active in my classroom fits well with the way I like to teach” (compatibility), “It will be difficult for me to learn how to provide opportunities for children to be physically active in my classroom” (complexity), “It is okay for me to try providing opportunities for children to be physically active in my classroom on a limited basis before fully implementing it in my daily routine” (trialability), and “Administrators will be able to see the results of providing opportunities for children to be physically active in my classroom” (observability). We reverse coded the items measuring complexity so that results were indicative of CTs’ perceived simplicity, which was more easily interpretable in relation to the other perceived attributes. Domain-Specific Innovativeness. To assess ECTs’ domain-specific innovative-
ness in the school environment, we adapted a questionnaire from Goldsmith and Hofacker (1991), which includes six items and uses a five-point response scale with the anchors 1 = “Strongly disagree” and 5 = Strongly agree.” The measure has been internationally validated in consumer science research (Goldsmith, d’Hauteville, & Flynn, 1998). We made several modifications to the original instrument. Specifically, the target domain was changed to “education/the classroom,” negatively worded items were changed to positively worded items, the term “adopt” was incorporated into the items to be consistent with the purpose of the study, and the scale range was increased from five points to seven points to increase response variability. These modifications were made in accordance with feedback we obtained from one of the developers of the original instrument, who we asked to examine our measure for content validity. Like the original instrument, our questionnaire uses six items and the same anchor terminology. Example items are “In general, I am among the first of the classroom teachers at my school to adopt a new educational idea/classroom practice” and “If I learned that a new educational idea/classroom practice was available, I would be interested enough to adopt it.”
Self-Reported PAPAC. We developed a 6-item questionnaire for this study to assess the frequency of ECTs’ self-reported PAPAC. Recommended strategies for PAPAC from the Student Health and Fitness Act and the school wide physical activity promotion literature (e.g., Maeda & Murata, 2004; Rink et al., 2010; Webster et al., 2010) were drawn upon to develop the items, which asked participants how often they integrate physical activity opportunities into academic lessons (e.g., “How often do you integrate physical activity opportunities into math lessons?”), how often they use active warm-up routines (“How often do you facilitate physically active exercises, such as a “warm-up” routine, for students in your classroom (i.e., that is not part of a school wide activity)?”), and how often they use physical activity breaks between lessons (“How often do you incorporate movement breaks (e.g., active transitions, “energizers”) between lessons in your classroom?”). The questionnaire uses a 5-point response scale with the following response options: “Never,” “Rarely,” “Sometimes,” “Often,” and “Very Often.”
Adoption of Physical Activity Promotion 427
Procedure Before collecting data for this study, we obtained approval to conduct the study from the institutional review board for research with human subjects at the University of South Carolina. The data reported in this manuscript are part of a larger study on ECTs’ PAPAC, which was supported by a grant from the American Alliance for Health, Physical Education, Recreation and Dance (AAHPERD) Research Consortium. We pilot tested the questionnaires with a different sample of 32 ECTs at a state funded elementary school in South Carolina. Pilot participants were asked to complete the questionnaire, make written and/or verbal suggestions for clarity and concision, and recommend changes to content that would more accurately reflect their viewpoints concerning PAPAC. Based on participants’ feedback, we changed the wording for several of the items to increase clarity and concision. To help ensure content validity, we asked the authors of the papers describing the development of the original diffusion of innovations measures adapted for this study to review our instrumentation and provide feedback about the questionnaires’ measurement properties and content. As stated earlier, one of the researchers who developed the original domain-specific innovativeness measure helped us in this regard. We also solicited feedback from five graduate students and three professors with backgrounds in school-based physical activity promotion. Their feedback served to corroborate the content of the PAPAC measure and was used to refine the organization and formatting of the questionnaires for readability. The final questionnaires were administered, in paper and pencil format, first to ECTs at a 2011 professional conference for elementary teachers in South Carolina (n = 113), and then to ECTs at four schools in one South Carolina school district (n = 88) to maximize the sample size. The teachers gave their consent to participate by completing the questionnaires. The incentive of winning an iPad2 was used to recruit ECTs attending the conference as participants in the study. All prospective participants were screened to make sure they were generalist classroom teachers who currently taught elementary schools in South Carolina and had not participated in the focus groups or the pilot study. Twelve participants reported having no South Carolina teaching experience and we excluded them from the sample. At each of the four schools, the school principal gave us permission to administer the questionnaires with ECTs during a regularly scheduled staff meeting. The same instructions were given, both verbally and in writing, to all participants in the study. Participants were told that their participation was voluntary, choosing not to participate would have no negative consequences, there were no right or wrong responses, and their responses would only be reported in aggregate with responses from other participants. It was stressed that no identifying information would be used (e.g., school affiliation) when reporting the data and participants were asked not to put their names on any of the questionnaires to help ensure confidentiality. Instructions were given for participants to respond to all items and to respond honestly, and to pay particular attention to the definition and examples of physical activity provided on the questionnaire packet before responding to any of the questionnaire items. Physical activity was defined as “behavior that substantially increases energy expenditure when compared to a resting state, such as sitting or lying down” (Pate, O’Neill, & Lobelo, 2002). The examples of physical activity given were “brisk walking, running, climbing, jumping, hopping, dancing, playing catch, and doing sit-ups or other body resistance exercises” (USDHHS, 2008).
428 Webster et al.
Participants took approximately 20 min to complete the full packet of questionnaires, including measures not reported in the current study.
Data Analysis Statistical analyses were conducted in SPSS (19.0) and SmartPLS (Ringle, Wende, & Will, 2005). Responses to all negatively stated items were reverse coded. The ShapiroWilk test in SPSS was used to examine the data for normality. The results of this test showed that the responses for each variable were not normally distributed. Based on this finding, and because our sample size was small, partial least squares analysis in SmartPLS was used to test the psychometric properties of our measures and the hypothesized relationships between variables. Sobel’s test was used to determine the statistical significance of indirect associations, including the relationships between policy awareness and intrapersonal variables and the relationship between perceived school support and PAPAC. The SPSS macro PROCESS was used to test the indirect relationship between policy awareness and PAPAC through all other variables in the model (Hayes, in press). This approach tests the indirect effects of mediators in serial and uses bootstrapping to estimate parameters and 95% bias corrected confidence intervals (CI). Means and standard variations for all variables were calculated using SPSS. Given we did not use a consistent sampling procedure for all participants in this study, we used bivariate correlations and partial correlations in SPSS to examine whether controlling for where participants completed the questionnaires (at the conference versus at their schools) changed the relationships between variables. The level of statistical significance did not change for any of the correlations, indicating that item responses did not vary based on where participants completed the questionnaires. Thus, the difference in where ECTs completed the instrumentation was not believed to be a factor in their perceptions concerning PAPAC.
Results Preliminary Psychometric Evaluation Means, standard deviations, and reliability coefficients are presented in Table 1. The reliability coefficients show that most of the measures are internally consistent. The exception is the trialability measure, which yielded a low Cronbach’s alpha coefficient, and a low to moderate composite reliability coefficient. We retained this measure in the model despite these low values, since the scores were not used to make judgments about individuals (Aiken, 1997). The construct validity of the measures was assessed by considering the item loadings for each measure (Table 2). The item loadings are greater than .4 with each item loading on its respective latent variable, thus providing support for the construct validity of the measures and the measurement component of the proposed path model (Tenenhaus, Vinzi, Chatelin, & Lauro, 2005). The discriminant validity of the measures was assessed by considering whether the square root of the average variance extracted (AVE) exceeds the intercorrelations between each given latent variable and the other latent variables in the research model (Fornell & Larcker, 1981). The latent variable correlations and square roots of the AVE are presented in Table 3, indicating that all measures meet the criterion for discriminant validity.
Table 1 Descriptive Statistics for Research Variables Measure
Mean
SD
Cronbach’s Alpha
Composite Reliability
Policy Awareness
2.62
.93
-
-
Perceived School Support
4.53
1.33
.86
.89
Relative Advantage
4.69
1.33
.51
.78
Compatibility
5.26
1.31
.86
.92
Simplicity
4.93
1.38
.78
.86
Trialability
5.40
.97
.40
.59
Observability
4.99
1.23
.76
.86
Domain-Specific Innovativeness
4.58
1.12
.84
.88
Self-Reported PAPAC
2.11
.48
.86
.90
Note. PAPAC = Physical Activity Promotion in the Academic Classroom. Scale ranges were 1–4 for Policy Awareness, 1–7 for Perceived School Support, 1–7 for Perceived Attributes (Relative Advantage, Compatibility, Simplicity, Trialability, and Observability), 1–7 for Domain-Specific Innovativeness, and 0–4 for Self-Reported PAPAC.
Table 2 Item Loadings for Measures Measure
Item Loading
Policy Awareness How much did you already know about this policy?
1.00
Perceived School Support My classroom environment can be easily modified to provide opportunities for my children to be physically active.
.74
Overall, my school administration poses a barrier to providing opportunities for children to be physically active in my classroom.
.51
The school schedule allows me to provide children with opportunities to be physically active in my classroom.
.82
I have enough materials to provide opportunities for children to be physically active in my classroom.
.71
There are too many obstacles in my classroom for me to safely provide opportunities for children to be physically active.
.65
The academic curriculum I follow makes it hard for me to provide children with opportunities to be physically active in my classroom.
.77
My students make it easy for me to provide them with opportunities to be physically active in my classroom.
.76
I do not have enough space in my classroom to provide opportunities for children to be physically active.
.69
429
Measure
Item Loading
Relative Advantage Providing opportunities for children to be physically active in my classroom will increase the quality of education my students receive.
.90
Providing opportunities for children to be physically active in my classroom will enhance my effectiveness as a classroom teacher.
.91
My students’ academic performance will decrease if I provide opportunities for them to be physically active in my classroom.
.46
Providing opportunities for children to be physically active in my classroom will increase my students’ ability to learn.
.86
Compatibility Providing opportunities for children to be physically active in my classroom fits well with the way I like to teach.
.86
Providing opportunities for children to be physically active in my classroom is consistent with my priorities as a teacher.
.90
Providing opportunities for children to be physically active in my classroom is compatible with my educational philosophy.
.89
Complexity (Reverse Coded to “Simplicity”) It will be difficult for me to learn how to provide opportunities for children to be physically active in my classroom.
.72
Providing opportunities for children to be physically active in my classroom would require me to make substantial changes to my teaching routines.
.77
Overall, it will be complicated for me to implement physical activity opportunities for children in my classroom.
.89
Providing opportunities for children to be physically active in my classroom requires more work than can be accomplished with current resources available to me.
.70
Trialability It is okay for me to try providing opportunities for children to be physically active in my classroom on a limited basis before fully implementing it in my daily routine.
.44
I am allowed to experiment with new ways to implement physical activity opportunities in my classroom.
.89
I can integrate physical activity opportunities for children in my classroom at my own pace.
.88
I am required to fully integrate physical activity opportunities for children into my daily routine as a classroom teacher right away.
.49
Observability Administrators will be able to see the results of providing opportunities for children to be physically active in my classroom.
.85 (continued)
430
Adoption of Physical Activity Promotion 431
Table 2 (continued) Measure
Item Loading
Observability (continued) Parents will not be able to see the results of providing opportunities for children to be physically active in my classroom.
.78
Other teachers at my school will be able to see the results of my providing opportunities for children to be physically active in my classroom.
.82
Domain-Specific Innovativeness In general, I am among the first of the classroom teachers at my school to adopt a new educational idea/classroom practice.
.70
If I learned that a new educational idea/classroom practice was available, I would be interested enough to adopt it.
.76
I adopt more new educational ideas/classroom practices than other classroom teachers at my school.
.83
I will consider adopting a new educational idea/classroom practice, even if it is the first time I have ever heard of it.
.60
In general, I am among the first of the classroom teachers at my school to know the latest trends in education/classroom teaching.
.80
I know more about new educational ideas/classroom practices before most other classroom teachers at my school do.
.76
Self-Reported PAPAC How often do you integrate physical activity opportunities into math lessons?
.79
How often do you integrate physical activity opportunities into language arts lessons?
.81
How often do you integrate physical activity opportunities into science lessons?
.78
How often do you integrate physical activity opportunities into social studies lessons?
.78
How often do facilitate physically active exercises, such as a “warm-up” routine, for students in your classroom (i.e., that is not part of a school wide activity)?
.74
How often do you incorporate movement breaks (e.g., active transitions, “energizers”) between lessons in your classroom?
.73
Model Testing The results of testing the various paths in the research model are presented in Table 4. We obtained estimates of the path coefficients and the associated t-statistics by bootstrapping 1000 resamples. All relationships hypothesized in the research model were supported, except for the path between relative advantage and selfreported PAPAC, and the path between trialability and self-reported PAPAC. The
432 Webster et al.
Table 3 Latent Variable Correlations 1. 1. Policy Awareness
1.00
2. Perceived School Support
.09
2.
3.
4.
5.
6.
7.
8.
9.
.71
3. Relative Advantage
.17
.56 .80
4. Compatibility
.10
.61 .75 .88
5. Simplicity
.08
.59 .48 .45 .78
6. Trialability
.20
.53 .53 .66 .33 .70
7. Observability
.15
.58 .75 .59 .42 .49 .82
8. Domain-Specific Innovativeness
.09
.35 .45 .46 .24 .41 .45 .74
9. Self-Reported PAPAC
.16
.56 .52 .64 .45 .47 .50 .45 .77
Note. The square root of the AVE is reported in the diagonal. PAPAC = Physical Activity Promotion in the Academic Classroom.
R2 values show that policy awareness explained 2% of the variance in perceived school support, which in turn explained 12% of the variance in domain-specific innovativeness, 32% of the variance in perceived relative advantage, 37% of the variance in perceived compatibility, 35% of the variance in perceived simplicity, 28% of the variance in perceived trialiability, and 34% of the variance in perceived Table 4 Path Coefficients and t-Statistics for Paths in the Research Model beta
SE
t
p
Policy Awareness -> Perceived School Support
0.15
0.05
3.14
0.002
Perceived School Support -> Relative Advantage
0.57
0.03
21.69
0.000
Perceived School Support -> Compatibility
0.61
0.03
24.42
0.000
Perceived School Support -> Simplicity
0.59
0.04
16.42
0.000
Perceived School Support -> Trialability
0.53
0.04
14.85
0.000
Perceived School Support -> Observability
0.58
0.03
20.29
0.000
Innovativeness
0.35
0.04
9.89
0.000
Relative Advantage -> PAPAC
-0.12
0.08
1.48
0.139
Compatibility -> PAPAC
0.44
0.08
5.82
0.000
Simplicity -> PAPAC
0.20
0.04
4.89
0.000
Observability -> PAPAC
0.16
0.05
3.01
0.003
Trialability-> PAPAC
0.03
0.04
0.61
0.544
Domain-Specific Innovativeness -> PAPAC
0.16
0.04
3.79
0.000
Perceived School Support -> Domain-Specific
Note. SE = Standard Error. PAPAC = Physical Activity Promotion in the Academic Classroom.
Adoption of Physical Activity Promotion 433
observability. Domain-specific innovativeness and the five perceived attributes explain 48% of the variance in self-reported PAPAC. While the R2 for perceived school support and domain-specific innovativeness are relatively low, the variances explained for the other latent variables are substantial (Cohen, 1988). Tenenhaus et al. (2005) proposed a statistic to assess the global fit of a PLS model. Their criterion of goodness of fit is defined as the squared root of product of the average communality and the average R2. Using this statistic, a global goodness of fit value of 0.416 was obtained for the model presented in this paper, suggesting that approximately 42% of the achievable fit is accounted for in this model. Schepers, Wetzels, and de Ruyter (2005) argued that a criterion for a large effect using this approach would be .36. On this basis, the goodness of fit measure would suggest a reasonable model fit to the data. Indirect relationships between variables are presented in Table 5. The results show that perceived school support mediates the relationship between policy awareness, each of the five perceived attributes, and domain-specific innovativeness. However, the indirect relationship between perceived school support and PAPAC is not mediated by relative advantage or trialability. The results of testing the indirect relationship between policy awareness and PAPAC through all other variables in the model are presented in Table 6. An examination of the CIs suggests that the indirect relationship between policy awareness and PAPAC through perceived school support and trialabiltiy is the only nonsignificant result (the interval contains zero). However, the other indirect relationships shown in the table have lower bound values that are very close to zero. Thus, the significant results should be interpreted with some caution, as they suggest only borderline significance. Table 5 Indirect Relationships Using Sobel’s Test Indirect path
Value
Sobel Statistic
p
Policy Awareness -> Domain-Specific Innovativeness
.053
2.99
.003
Policy Awareness -> Relative Advantage
.086
3.11
.002
Policy Awareness -> Compatibility
.092
3.12
.002
Policy Awareness -> Simplicity
.089
3.08
.002
Policy Awareness -> Trialability
.078
3.07
.002
Policy Awareness -> Observability
.087
3.10
.002
Perceived School Support -> PAPAC (Domain-Specific Innovativeness)
.056
3.54
.000
Perceived School Support -> PAPAC (Relative Advantage)
-.068
1.48
.140
Perceived School Support -> PAPAC (Compatibility)
.268
5.66
.000
Perceived School Support -> PAPAC (Simplicity)
.118
4.69
.000
Perceived School Support -> PAPAC (Trialability)
.016
.61
.542
Perceived School Support -> PAPAC (Observability)
.093
3.01
.003
Note. PAPAC = Physical Activity Promotion in the Academic Classroom.
434 Webster et al.
Table 6 Indirect Effect of Policy Awareness on PAPAC Effect
SE
95% LCI
95% UCI
Perceived School Support -> Relative Advantage
Indirect path
.0175
.0100
.0001
.0402
Perceived School Support -> Compatibility
.0352
.0194
.0005
.0773
Perceived School Support -> Simplicity
.0134
.0106
.0002
.0442
Perceived School Support -> Trialability
.0024
.0027
-.0005
.0121
Perceived School Support -> Observability
.0208
.0124
.0013
.0512
Perceived School Support -> Domain-Specific Innovativeness
.0106
.0072
.0004
.0296
Note. PAPAC = Physical Activity Promotion in the Academic Classroom. SE = Standard Error. LCI = Lower Confidence Interval. UCI = Upper Confidence Interval.
Discussion PAPAC can be an effective way to increase elementary children’s physical activity at school (e.g., Bartholomew & Jowers, 2011; Erwin et al., 2011; Holt et al., 2012). Yet, despite a preponderance of recommended strategies and programs to increase children’s classroom-based physical activity (Faber, Kulinna, & Darst, 2007; Kahan, 2008; Maeda & Murata, 2004; NASPE, no date), little research has examined factors related to ECTs’ PAPAC adoption. In the current study, we drew from diffusion of innovations theory (Rogers, 1995) and a social ecological perspective (McLeroy et al., 1988; Spence & Lee, 2003) to test a theoretical model of South Carolina ECTs’ PAPAC adoption in the context of a statewide policy. Research exploring such factors is an important step in translating policy to practice, and developing effective PAPAC training and interventions. Given the variables targeted in this study have not previously been examined in relation to ECTs’ PAPAC, it was necessary to first test the strength of our measures. Overall, the psychometric evaluation showed that the items we used to define the variables were representative of the underlying constructs. Factor loadings fell where we expected them to, indicating good construct validity. Moreover, the criterion for discriminant validity was satisfied, indicating multicollinearity was not an issue. The reliabilities were within an acceptable range for all variables with only two exceptions (relative advantage and trialability). Previous school-based health promotion research has also found these variables to be problematic (Dechesnes et al., 2009; Pankratz et al., 2002). As stated earlier, Pankratz et al. (2002) eliminated their three items for trialability due conceptual problems and low internal consistency (.31). In addition, in a study examining school officials’ perceived attributes of a healthpromoting school innovation, Deschesnes et al. (2009) reported moderate internal consistency for their two relative advantage items (.60). Despite the steps taken in the current study to improve upon Pankratz et al.’s (2002) work (e.g., attempting to more closely align items with the established definitions of each variable, writing more items), there appear to still be conceptual issues that require further exploration. The results of model testing support the assertion that our theoretical framework for this study was quite robust. Most of the hypothesized direct relationships were
Adoption of Physical Activity Promotion 435
confirmed in the path model, as shown by the significant t values for all variables except relative advantage and trialability. The variances explained in perceived attributes by perceived school support and the variance explained in PAPAC by perceived attributes and domain-specific innovativeness are substantial. Based on these results, we suggest ECTs who perceive the school environment as more supportive of PAPAC will be more likely to view PAPAC favorably in terms of its relative advantage, compatibility, simplicity, trialability, and observability. Furthermore, ECTs who view PAPAC as compatible with their teaching philosophies and skills, as simple to adopt, and as capable of producing observable results to important members of the school, and who view themselves as innovative educators, will be more likely to adopt PAPAC (i.e., use PAPAC strategies frequently). Though statistically significant, awareness of the state policy explained little variance in perceived school support. Therefore, this study provides only tenuous evidence for a direct relationship between these variables. It is possible that a stronger relationship between these variables might have emerged if we had either expanded our policy awareness measure to include items focused on PAPAC strategies, or changed our perceived school support measure to focus more generally on support for physical activity promotion rather than specifically on PAPAC. As with the direct relationships, most of the hypothesized indirect relationships were supported. These results showed that the relationship between policy awareness and intrapersonal variables relies on the presence of school support. In other words, ECTs who were aware of the policy were more likely to perceive PAPAC favorably and feel innovative as educators when they perceived the school environment as supportive of PAPAC. The results also showed that ECTs who perceived the school environment as supportive of PAPAC were more likely to have adopted PAPAC when they perceived PAPAC as compatible, simple, and observable, and perceived themselves as innovative educators. The indirect relationships between policy awareness and self-reported PAPAC, mediated through the other variables in the model, were in most cases statistically significant, but perhaps not practically meaningful. The Student Health and Fitness Act may have had some influence on ECTs’ PAPAC adoption but the school environment and, in particular, intrapersonal variables, appear to have been much more important in the adoption process. The global goodness of fit estimate for the path model was adequate but would likely have improved if the relationship between relative advantage and PAPAC and the relationship between trialability and PAPAC had been supported. The low reliabilities for relative advantage and trialability affect the variability of the scores (item responses) for these variables, which subsequently affects the correlations in the model. There may also be other reasons these variables were not influential in ECTs’ PAPAC. While relative advantage has generally been a strong predictor of adoption in diffusion studies (Rogers, 1995), Deschesnes et al. (2009) found it was not a reliable predictor of adoption of a health promoting schools innovation (R2 = .02 using logistic regression). Rogers (2003) makes a distinction between incremental and preventive interventions. On one hand, incremental interventions provide beneficial outcomes in the short term and have a relatively short adoption period because the relative advantage is viewed with a high degree of certainty. Much of the previous research examining perceived attributes was conducted with agricultural interventions, which often quickly yielded benefits and could be characterized as incremental (Rogers, 1995). On the other hand, preventive interventions provide
436 Webster et al.
beneficial outcomes in the long term and the relative advantage (i.e., the rewards of the adoption) is therefore perceived as uncertain. School-based health promotion innovations might not be expected to produce results as quickly as farming innovations and could therefore be more appropriately characterized as preventive. Another reason for the nonsignificant relationship between relative advantage and PAPAC in this study could be that the ECTs did not perceive the benefits reflected in the relative advantage items as the most important benefits of PAPAC. The items focused on increasing the quality of education that students receive, enhancing the teacher’s effectiveness, increasing students’ academic performance, and increasing students’ ability to learn. Based on the prevalent accountability culture focused on students’ academic achievement in schools, our discussions with ECTs in the focus group interviews, and the results of our pilot testing, we assumed ECTs would primarily evaluate the merits of PAPAC in accordance with these possible advantages. However, Rogers (1995) specifies six subdimensions of relative advantage, including the degree of economic profitability, low initial cost, decrease in discomfort, social prestige, savings in time and effort, and immediacy of the reward. It is possible that items drawing more fully from these different subdimensions might tap into advantages that are more in line with ECTs’ priorities. For instance, an important economic advantage of PAPAC for teachers that was not assessed in this study could be the “reduced cost”, or having to work less to achieve the same or better results from students, such as on-task behavior. Relative advantage might also be better assessed if teachers were able to compare the innovation to a concrete alternative (i.e., “Teaching physically active classroom lessons will increase my students’ academic performance more than teaching sedentary lessons with students at their desks”). Trialability has received relatively little attention in health promotion research (Pankratz et al., 2002). Pankratz et al. (2002) suggest that this could be because the construct is difficult to measure or it has not been seen as significant in health promotion research. Dechesnes et al. (2009) did not consider trialability relevant in their study of a health-promoting schools intervention and did not include the construct in their measure of perceived attributes. Trialability could be easier to assess and have more relevance to potential health promotion adopters when the innovation is defined in more concrete terms than the one investigated in the current study (Pankratz et al., 2002). PAPAC consists of numerous strategies; focusing on one particular strategy (e.g., infusing physical activity breaks between academic lessons) might improve the measurability, and increase the influence, of trialability. In addition, some teachers might perceive PAPAC as trialable but feel the results of previous trials were negative and, therefore, report using PAPAC strategies infrequently. This study had several limitations. Though drawn from a broad geographic region of South Carolina, the sample was not randomly selected and only 12% of all schools in the state were represented. This limits the generalizability of the findings. In addition, the use of measures with different response scales could have led to variation in responses based on scale range. From a diffusion of innovations perspective, our measures of relative advantage and trialability may not have been sufficiently contextualized. Continued research examining PAPAC will benefit from elicitation studies with ECTs to generate the most cogent qualities to focus on when developing and adapting perceived attributes scales for use in school-based health
Adoption of Physical Activity Promotion 437
promotion research (Rogers, 1995). From a social ecological perspective, we did not investigate the possibility of reciprocal influences within our model (Bronfenbrenner, 1977; Langille & Rodgers, 2010). For instance, PAPAC adoption might influence school support, as teachers could begin to champion PAPAC and garner more resources for it. Investigating such “upstream” influences would extend and refine theoretical perspectives applied to PAPAC training and interventions. Future research might also consider targeting groups of individuals or the organization (i.e., school) as the unit of analysis. McLeroy et al. (1988) recommends that social ecological researchers move beyond focusing on “changing individuals through social influences [to] changing the norms or social groups to which individuals belong” (p. 357). This shift in focus could help to increase the impact of schoolbased interventions. In conclusion, this study provides several measures that can be adapted in further research examining PAPAC, and demonstrates the importance of policy awareness, a supportive school environment, perceived attributes, and domainspecific innovativeness in ECTs’ PAPAC adoption. Based on the results of this study, efforts to increase ECTs’ PAPAC adoption should focus on: (a) increasing school support for PAPAC (e.g., principal buy-in, sufficient classroom space, provision of relevant resources/materials), (b) increasing ECTs’ perceptions that PAPAC is compatible with current educational practice at the school, simple to carry out, and observable as part of a successful school program, and (c) rewarding ECTs for trying new educational ideas and practices so as to nurture domain-specific innovativeness. Since current conceptual models of school wide physical activity (e.g., the Comprehensive School Physical Activity Program; NASPE, 2008) view the leadership of physical educators as central to the expansion of physical activity opportunities across school contexts, the information from this study has direct relevance to physical education professionals. Those in the physical education field can draw from this information to target key mechanisms for, and pursue appropriate channels to, increasing PAPAC adoption. As policy initiatives for school wide physical activity unfurl, it is imperative that research continues to identify variables related to the adoption process. Only once a clear understanding of the adoption process emerges can a pathway to effective diffusion become a realistic pursuit (Metzler et al., 2008).
References Aiken, L.R. (1997). Psychological testing and assessment (9th ed.). Boston, MA: Allyn and Bacon. Bartholomew, J.B., & Jowers, E.M. (2011). Physically active academic lessons in elementary children. Preventive Medicine, 52, 551–554. doi:10.1016/j.ypmed.2011.01.017 Bass, F.M. (1969). A new product growth model for consumer durables. Management Science, 15(5), 215–227. doi:10.1287/mnsc.15.5.215 Beets, M.W., Flay, B.R., Buchinich, S., Acock, A.C., Li, K., & Allred, C. (2008). School climate and teachers’ beliefs and attitudes associated with implementation of the positive action program: A diffusion of innovations model. Prevention Science, 9(4), 264–275. doi:10.1007/s11121-008-0100-2 Bronfenbrenner, U. (1977). Toward an experimental ecology of human development. The American Psychologist, 32, 513–531. doi:10.1037/0003-066X.32.7.513
438 Webster et al.
Bronfenbrenner, U. (1979). The ecology of human development. Cambridge, MA: Harvard University Press. Citrin, A.V., Sprott, D.E., Silverman, S.N., & Stem, D.E. (2000). Adoption of Internet shopping: The role of consumer innovativeness. Industrial Management & Data Systems, 100(7), 294–300. doi:10.1108/02635570010304806 Cohen, J. (1988). Statistical power analysis for the behavioural sciences. Hillsdale, NJ: Lawrence Erlbaum Associates. Craig, S.C., & Ginter, J.L. (1975). An empirical test of a scale for innovativeness. In M.J. Schlinger (Ed.), Advances in consumer research (Vol. 2, pp. 555–562). Chicago, IL: Chicago Association for Consumer Research. Dearing, J.W. (2009). Applying diffusion of innovation theory to intervention development. Research on Social Work Practice, 19(5), 503–518. doi:10.1177/1049731509335569 Deschesnes, M., Trudeau, F., & Kébé, M. (2009). Psychometric properties of a scale focusing on perceived attributes of a health promoting school approach. Canadian Journal of Public Health, 100(5), 389–392. Erwin, H.E., Beighle, A., Morgan, C.F., & Noland, M. (2011). Effect of a low-cost, teacherdirected classroom intervention on elementary students’ physical activity. The Journal of School Health, 81(8), 455–461. doi:10.1111/j.1746-1561.2011.00614.x Erwin, H.E., Fedewa, A., Beighle, A., & Ahn, S. (2012). A quantitative review of physical activity, health, and learning outcomes associated with classroom-based physical activity interventions. Journal of Applied School Psychology, 28(1), 14–36. doi:10.10 80/15377903.2012.643755 Faber, L., Kulinna, P.H., & Darst, P.W. (2007). Strategies for physical activity promotion beyond the physical education classroom. Journal of Health, Physical Education, Recreation, and Dance, 78(9), 27–31. doi:10.1080/07303084.2007.10598095 Fornell, C.R., & Larcker, D.F. (1981). Structural equation models with unobservable variables and measurement error. JMR, Journal of Marketing Research, 18(3), 39–50. doi:10.2307/3151312 Gatignon, H., & Robertson, T.S. (1985). A propositional inventory for new diffusion research. The Journal of Consumer Research, 11, 849–867. doi:10.1086/209021 Goldsmith, R.E., d’Hauteville, F., & Flynn, L.R. (1998). Theory and measurement of consumer innovativeness: A transnational evaluation. European Journal of Marketing, 32(3), 340–353. doi:10.1108/03090569810204634 Goldsmith, R.E., & Hofacker, C.F. (1991). Measuring consumer innovativeness. Journal of the Academy of Marketing Science, 19(3), 209–221. doi:10.1007/BF02726497 Hall, T.J., Little, S., & Heidorn, B.D. (2011). Preparing classroom teachers to meet students’ physical activity needs. Journal of Physical Education, Recreation & Dance, 82(3), 40–52. doi:10.1080/07303084.2011.10598596 Hayes, A.F. (in press). An introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press. Hirschman, E.C. (1980). Black ethnicity and innovative communication. Journal of the Academy of Marketing Science, 8, 100–118. doi:10.1007/BF02721977 Ho, C., & Wu, W. (2011). Role of innovativeness of consumer in relationship between perceived attributes of new products and intention to adopt. International Journal of Electronic Business Management, 9(3), 258–266. Holt, E., Heelan, K., & Bartee, T. (2012). Evaluation of a policy to integrate physical activity into the school day. Journal of Physical Activity and Health. Joseph, B., & Vyas, S.J. (1984). Concurrent validity of a measure of innovative cognitive style. Journal of the Academy of Marketing Science, 12(2), 159–175. doi:10.1007/ BF02729494 Kahan, D. (2008). Fitting physical activity into the elementary school day. Leadership Compass, 6(1), 1–3.
Adoption of Physical Activity Promotion 439
Kreuger, R., & Casey, M.A. (2000). Focus groups: A practical guide for applied research (3rd ed.). Thousand Oaks, CA: Sage. Langille, J.D., & Rodgers, W.M. (2010). Exploring the influence of a social ecological model on school-based physical activity. Health Education & Behavior, 37(6), 879–894. doi:10.1177/1090198110367877 Lochman, J.E. (2003). Commentary: School contextual influences on the dissemination of interventions. School Psychology Review, 32, 174–177. Maeda, J.K., & Murata, N.M. (2004). Collaborating with classroom teachers to increase daily physical activity: The Gear Program. Journal of Physical Education, Recreation & Dance, 75(5), 42–46. doi:10.1080/07303084.2004.10607239 Mahar, M.T., Murphy, S.K., Rowe, D.A., Golden, J., Shields, T., & Raedeke, T.D. (2006). Effects of a classroom-based program on physical activity and on-task behavior. Medicine and Science in Sports and Exercise, 38(12), 2086–2094. doi:10.1249/01. mss.0000235359.16685.a3 McKenzie, T.L., Sallis, J.F., & Rosengard, P. (2009). Beyond the stucco tower: Design, development, and dissemination of the SPARK physical education programs. Quest, 61, 114–127. doi:10.1080/00336297.2009.10483606 McLeroy, K.R., Bibeau, D., Steckler, A., & Glanz, K. (1988). An ecological perspective on health promotion programs. Health Education Quarterly, 15, 351–377. doi:10.1177/109019818801500401 Metzler, M.W., Lund, J.L., & Gurvitch, R. (2008). Chapter 2: Adoption of instructional innovation across teachers’ career stages. Journal of Teaching in Physical Education, 27, 457–465. National Association for Sport and Physical Education. (2008). Comprehensive school physical activity programs. Retrieved November 13, 2012 from http://www.aahperd. org/naspe/pdf_ files/cspaponline.pdf. National Association for Sport and Physical Education. (no date). Integrating physical activity into the complete school day. Retrieved December 10, 2012 from http://www. aahperd.org/naspe. Owen, N., Glanz, K., Sallis, J.F., & Kelder, S.H. (2006). Evidence-based approaches to dissemination and diffusion of physical activity interventions. American Journal of Preventive Medicine, 31(4S), S35–S44. doi:10.1016/j.amepre.2006.06.008 Pankratz, M., Hallfors, D., & Cho, H. (2002). Measuring perceptions of innovation adoption: The diffusion of a federal drug prevention policy. Health Education Research, 17(3), 315–326. doi:10.1093/her/17.3.315 Pate, R.R., O’Neill, J.R., & Lobelo, F. (2002). The evolving definition of “sedentary.”. Exercise and Sport Sciences Reviews, 36, 173–178. doi:10.1097/JES.0b013e3181877d1a Ringeisen, H., Henderson, K., & Hoagwood, K. (2003). Context matters: schools and the ‘‘research to practice gap’’ in children’s mental health. School Psychology Review, 32, 153–168. Ringle, C.M., Wende, S., & Will, S. (2005). SmartPLS 2.0 (M3) (Beta). Hamburg. Retrieved November 14, 2012 from http://www.smartpls.de. Rink, J.E., Hall, T.J., & Williams, L.H. (2010). Schoolwide physical activity: A comprehensive guide to designing and conducting programs. Champaign, IL: Human Kinetics. Rogers, E.M. (1995). Diffusion of innovations (4th ed.). New York, NY: Free Press. Rogers, E.M. (2003). Diffusion of innovations (5th ed.). New York, NY: Free Press. Rogers, E.M., & Shoemaker, F.F. (1971). Communication of innovations. New York, NY: The Free Press. Schepers, J., Wetzels, M., & de Ruyter, K. (2005). Leadership styles in technology acceptance: Do followers practice what they preach? Managing Service Quality, 15, 496–508. doi:10.1108/09604520510633998 South Carolina Student Health and Fitness Act of 2005. (2005). Retrieved November 13, 2012, from http://www.scstatehouse.gov/sess116_2005-2006/bills/3499.htm.
440 Webster et al.
Spence, J.C., & Lee, R.E. (2003). Toward a comprehensive model of physical activity. Psychology of Sport and Exercise, 4(1), 7–24. doi:10.1016/S1469-0292(02)00014-6 Tenenhaus, M., Vinzi, V.E., Chatelin, Y., & Lauro, C. (2005). PLS path modeling. Computational Statistics & Data Analysis, 48(1), 159–205. doi:10.1016/j.csda.2004.03.005 Thomas, J.R., Nelson, J.K., & Silverman, S.J. (2011). Research methods in physical activity (6th ed.). Champaign, IL: Human Kinetics. U.S. Department of Health and Human Services. (2001). The Surgeon General’s call to action to prevent and decrease overweight and obesity. Retrieved November 15, 2012, from http:// www.surgeongeneral.gov/library/calls/obesity/index.html. U.S. Department of Health and Human Services. (2008). Physical activity guidelines for Americans. Washington, DC: U.S. Department of Health and Human Services. US National Physical Activity Plan. Retrieved February 24, 2013 from www.physicalactivityplan.org. Webster, C.A. (2011). Relationships between personal biography and changes in preservice classroom teachers’ physical activity promotion competence and attitudes. Journal of Teaching in Physical Education, 30(4), 320–339. Webster, C.A., Erwin, H., & Parks, M. (in press).Relationships between and changes in preservice classroom teachers’ efficacy beliefs, willingness to integrate movement, and perceived barriers to movement integration. Physical Educator. Webster, C.A., Monsma, E., & Erwin, H. (2010). The role of biographical characteristics in preservice classroom teachers’ physical activity promotion attitudes. Journal of Teaching in Physical Education, 29, 358–377.
