Changes in exercise commitment following a values-based wellness program.

Brinthaupt, Thomas M. ; Kang, Minsoo ; Anshel, Mark H. 等

Despite the benefits of regular exercise and healthy eating, most people have difficulty beginning and maintaining positive behavioral changes. For example, large percentages of novice exercisers will eventually drop out of structured exercise or fitness programs (Dishman, 2001; Kelly & Warnick, 1999). Researchers and practitioners have examined the effects of various wellness programs on health behavior change. However, interventions that provide information and instruction intended to improve dietary and exercise habits have been only moderately successful (Marcus, Ciccolo, Whitehead, King, & Bock, 2009).

Health behavior change is difficult, in part due to people's lack of comfort and familiarity with using self-regulatory behaviors (Ockene, 2001). Winett, Anderson, Wojcik, Winett, and Bowden (2007) argued proper self-regulation skills (e.g., goal-setting, self-monitoring, planning, time management) are necessary to produce small, sustained shifts in health behaviors. Thus, programs that focus on establishing and developing self-regulatory strategies are thought to be more effective at bringing about positive change than programs that ignore these strategies.

Researchers and practitioners contend that it is more desirable to use behavior change interventions that are self-generated and self-determined than externally imposed efforts to change health behaviors, which are extrinsically motivated and less desirable (Rollnick, Mason, & Butler, 1999). According to self-determination theory (SDT; Deci & Ryan, 2002), interventions that foster autonomous motivation will be most effective. SDT defines autonomous or self-determined motivation as generally consisting of both identified regulation (i.e., individuals exercise or eat well because they think it is important or because they value good health) and integrated regulation (i.e., individuals perceive exercise or healthy eating as consistent with their goals, needs, and values). In addition, controlled motivation refers to engaging in health behaviors for external or introjected (e.g., shame or guilt) reasons, and amotivation refers to instances where a person lacks any purposeful intention or motivation.

A large body of research supports the utility and validity of SDT in the realms of sport, exercise, and wellness (for reviews, see Hagger & Chatzisarantis, 2007a, b; Ryan, Williams, Patrick, & Deci, 2009; Wilson, Mack, & Grattan, 2008). People who engage in exercise for more intrinsic than extrinsic reasons tend to exercise more often and more persistently (e.g., Frederick, 2002; Puente & Anshel, 2009). Thus, based on SDT, an important element of a comprehensive wellness program should be to foster more intrinsic, autonomous self-regulated behavior.

Another factor related to the psychological aspects of wellness is commitment. When making health behavior changes, high levels of commitment to beginning and maintaining those changes are likely to improve people's successes. If individuals strengthen their engagement with and dedication to exercising, this has the potential to make wellness interventions more effective. One well-known model of commitment in competitive sports (Scanlan, Carpenter, Schmidt, Simons, & Keeler, 1993) consists of four components: involvement opportunities, personal investment, enjoyment, and social constraints. In a study of health club participants, all four of these components were significant predictors of exercise commitment (Alexandris, Zahariadis, Tsorbatzoudis, & Grouios, 2002).

Relatively limited research has been conducted on the role of commitment in the realm of health behavior change. One study of university students in an exercise class found that having a functional resolve or wanting to commit to exercise was associated with self-reported increases in exercise frequency (Wilson et al., 2004). Other researchers have shown that such intrinsic or volitional reasons are related to greater exercise persistence (e.g., Mullen & Markland, 1997) and that individuals with a stronger commitment to and perceived behavioral control over their exercising report greater correspondence between their intentions to exercise and their actual exercise behavior (Rhodes & Matheson, 2005).

In summary, theorists and researchers have suggested that autonomous motives coupled with higher levels of commitment can improve health behavior change efforts. Intervention models have increasingly recognized the importance of facilitating the autonomy and strengthening the behavioral commitment of participants. In the current study, one such model is described and its effects on participants' reasons for and commitment to exercising and eating well were tested.

Overview of the Disconnected Values Model

The Disconnected Values Model (DVM) is a recently developed approach to health behavior change (Anshel, 2008; Anshel & Kang, 2007). The model assumes that a "disconnection" or misalignment between one's values and one's actions will inhibit change efforts. The costs of a negative habit, such as eating poorly or not exercising, are typically well known by those with that habit. Individuals who are aware of the long-term consequences of unhealthy behaviors are more likely to take health protective actions (Hall & Fong, 2003). The DVM encourages individuals to take a long-term time perspective on the perceived costs and benefits of their negative habits.

A wellness program based on the DVM has several major components. First, participants identify negative habits that compromise their overall wellness. Second, they determine the benefits, short-term costs, and long-term consequences of each negative habit. Third, participants decide whether one or more of the disconnects between their personal values and current behaviors are unacceptable, given the associated costs and consequences. Finally, participants develop an "action plan" to replace their unhealthy habits with more desirable, healthier routines and work with a health professional on implementing that plan.

In the DVM program, fitness and nutritional coaches have the important role of cultivating participants' feelings of competence and autonomy. The coaches address participants' commitment and self-regulation by developing, encouraging, and monitoring the implementation of their action plans. This approach has been shown to improve positive affect as well as exercise motivation and frequency (e.g., Gagne, Ryan, & Bargmann, 2003; Puente & Anshel, 2009).

Similar to the approach of Motivational Interviewing (Miller & Rolinick, 2002), the DVM highlights the individuals' perceived conflict between important personal goals and their current behavior patterns. The DVM, however, goes beyond the recognition of dissonance by asking individuals to acknowledge the costs and long-term consequences of their negative habits and to develop a self-regulation action plan that carries out cognitive-behavioral change strategies. At the DVM's core is the self-motivated drive to recognize the inconsistency between one's values and negative, self-destructive habits, and then to use an action plan to institute a new, positive habit that is strongly connected to those values (Anshel, 2008; Loehr & Schwartz, 2003). More details about the theoretical foundations of the model and its practical application can be found elsewhere (Anshel, 2008; Anshel & Kang, 2007).

Tests of the DVM have shown support for its efficacy. For example, participants in a comprehensive DVM wellness intervention showed significantly improved pretest to posttest fitness scores and significant reductions in the disconnected values of health and happiness (Anshel, Kang, & Brinthaupt, 2010). In addition to improved physical fitness scores, participants also reported significantly improved scores on mental well-being (chronic anxiety, depression, positive affect, self-control, general health, and vitality) following the program (Anshel, Brinthaupt, & Kang, 2010). Research also has shown reductions in several perceived exercise barriers after participating in a DVM-based intervention (Brinthaupt, Kang, & Anshel, 2010).

Despite previous demonstrations of its effectiveness, questions remain regarding whether the DVM fosters changes in participants' motivations for and commitment to positive health behaviors. The purpose of this study, therefore, was to determine whether participating in a comprehensive wellness program would be associated with stronger commitment to healthier habits, particularly habits with autonomous motivational aspects. Based on past research, improvements in fitness and dietary habits were expected to reflect more "internal" than "external" reasons among program participants. Because of the orientation of the DVM, program participants were expected to show improvements in autonomous motivation and reductions in amotivation. Participants also were expected to report greater commitment to their health behavior efforts following the program.

Method

Participants

Participants were full-time faculty and staff from a large public university in the southeastern U.S., who signed up for a wellness program designed for those who are "unfit" and non-exercisers. Because the program was subsidized by a university grant, participants paid a nominal $25 registration fee to help cover a program t-shirt, fitness testing, and their coaching. Poor fitness was confirmed in pre-testing. A total of 62 participants started the program, nine of whom dropped out. Based on follow-up interviews, reasons for dropping out of the program were mainly idiosyncratic. These reasons included not having enough time (n = 4), not being able to schedule exercise with one's coach (n = 2), not liking one's coach (n = 1), injury (n = I), and having a medical procedure that prevented exercise (n = 1). Although participants were offered a full refund within two weeks of starting the program, none of the drop-outs requested a refund of their registration fee.

The final sample left 53 participants (42 females, 11 males), who ranged in age from 24 to 66 yrs. (M = 48.13 yrs, SD = 9.67). Participants received pre- and post-intervention fitness testing and completed the other pre- and post-intervention measures. The university Institutional Review Board approved the study and informed consent was obtained from all participants. Participants obtained clearance from their physician to participate in the 10week program.

Measures and Equipment

Exercise Commitment Index. A self-report Exercise Commitment Index (ECI) was developed for this study to assess participants' motivation to exercise before and after the program. Items were adapted from two sources. One conceptual framework linked a person's values to developing new, healthier habits from a highly personal perspective (Groppel, 2000). Sample items from this source included "I am willing to sacrifice other things to improve my fitness" and "1 take personal responsibility for my health, fitness, and well-being."

The Sport Commitment Model (Alexandris et al., 2002; Scanlan et al., 1993) served as the second source for the ECI. In this model, commitment is defined "as a psychological state representing the desire or resolve to continue sport participations" (Scanlan et al., 1993; p. 1). Whereas the instrument based on this model consisted of items related to sport commitment, the items for the ECI ascertained the participants' commitment to maintaining their exercise routines. Sample items from this source included "I am willing to sacrifice other things to improve my fitness" and "I am determined to reach my exercise goals." Respondents rated the 17 ECI items in terms of how committed they were to improving their health and fitness through regular exercise, using a 5-point Likert-type scale (1 = very low, 5 = very high).

To assess the psychometric properties of the ECI, corrected item-total and Cronbach's alpha coefficients were analyzed. A minimum threshold value for the corrected item-total coefficient was set at. 19, with items below this threshold judged as problematic (Allen & Yen, 1979). An initial analysis for the pretest ECI showed that one item had a corrected item-total coefficient equal to. 15. Another item pertained to respondents' intentions to complete this specific wellness program and did not apply at the posttest. These items were removed from the ECI, resulting in a final version with 15 items, from which an average ECI score was calculated. Cronbach's alpha coefficients for the ECI were high for both pretest ([alpha] = .92) and posttest ([alpha] = .90). These results supported the internal consistency of the commitment measure.

Reasons for Eating a Healthier Diet and Reasons for Exercising. Two measures were used from the Treatment Self-Regulation Questionnaire (TSRQ) battery developed in support of self-determination theory (Self Determination Theory, n.d.). The TSRQ is part of a family of self-regulation measures associated with motivations for engaging in healthy behaviors. Each TSRQ scale contains 15 items pertaining to either autonomous motivation (engaging in behaviors for self-determined and intrinsic reasons), controlled motivation (engaging in behaviors for external or introjected reasons), or amotivation (lacking any purposeful intention or motivation). Items for each subscale are averaged to assess the motivation type for the specific behavioral realm. Respondents rate the extent to which each reason is true for them using a 7-point scale (1 = Not at all true, 7 = Very true). Research supports the reliability and validity data for these measures (Levesque et al., 2007).

The Treatment Self-Regulation Questionnaires for Eating a Healthier Diet (TSRQ-Diet) and for Exercising (TSRQ-Exercise) provide reasons respondents would either start eating a healthier diet [or exercise regularly] or would continue to do so. Each item is rated using the same stem--"The reason I would eat a healthy diet [or exercise regularly] is ..." An example autonomous motivation item was "because I personally believe it is the best thing for my health." A controlled motivation sample item was "because I would feel bad about myself if I did not eat a healthy diet [or exercise regularly]." An amotivation item was "I really don't think about it."

Fitness Tests. Participants completed four fitness tests within one week of the program's formal beginning, and again at the conclusion of the intervention, following the guidelines recommended by the American College of Sports Medicine (ACSM, 2006). Body composition was measured with a Lange skinfold caliper (Beta Technology Inc., Santa Cruz, CA), using a 7-site assessment (Pollock & Jackson, 1984). Cardiovascular fitness was assessed using the estimated V[O.sup.2] max Single-Stage Treadmill Test (Ebbeling, Ward, Puleo, Widrick, & Rippe, 1991). Participants completed this test on a Quinton Treadmill, Model number Q55, using standard protocol. Upper and lower body muscular strength was measured using a Universal Weight Machine (model no. SS1500, Universal Gym Equipment Co., West Point, MS). Participants were briefly instructed as to proper form and breathing technique before performing each test (ACSM, 2006). Bench press was used for upper body, and leg press was used for lower body testing. Participants then performed as many repetitions at the selected weight resistance until fatigue was established (i.e., the person could not complete another repetition), with a maximum of 15 repetitions.

Design and Procedure

The study was a one-group pretest/posttest design. The 10-week intervention began with a 3-hour group orientation based on guidelines and concepts from the DVM (see Anshel, 2008, and Anshel, Kang, et al., 2010 for additional details about the orientation and DVM). Prior to the orientation, participants completed a registration form on which they indicated their preference for a male or female fitness coach and their preferred exercise testing and workout schedule for meeting with their fitness coach (e.g., morning, afternoon, or evening). Before the orientation, participants were assigned to their coach based solely on these two criteria and met their coach at the program's orientation, at which time the DVM was reviewed.

Immediately following the orientation seminar, participants met their fitness coaches. Participants did not know their coaches prior to beginning the program. The fitness coaches (n = 6, two women, four men) were graduate students specializing in the university's exercise science program who had undergone a 3-hour training orientation with one of the researchers. Coaches ranged in age from 24 - 29 yrs (M = 27.2, SD = 2.02). They were assigned between seven and nine participants based on their available time and the participants' expressed preferred exercise times (e.g., early a.m., lunch time, evenings). Coaches were compensated at $15/hour for their work, which included administering pre- and post-intervention fitness tests. To be hired, each coach had to have completed a university graduate course in "fitness testing and prescription" and had previous experience as a personal trainer. In addition, all fitness coaches were required to complete a two-hour in-service training program led by an exercise science doctoral candidate from the campus program on proper fitness testing, prescription, and training for the middle-aged, unfit participants in this study.

Because the program was advertised to the campus community as a comprehensive wellness program, which included fitness and nutrition, we devoted some attention to the eating behaviors of the participants. Although it was not a primary purpose of the intervention, this element was included to provide participants with some general guidance and additional resources pertaining to their eating habits. Within the first two weeks of the program, participants met individually for 30 minutes with a registered dietician (RD) to discuss their past eating patterns, future nutritional needs, and possible dietary changes. No clients were assigned a particular diet. The RD also conducted weekly optional 45-minute group nutrition education seminars during the intervention period. All participants could access the nutrition-related educational materials through a program Website.

At the end of the program orientation, participants developed an individualized 24hour action plan that included the new daily or weekly routines that were incorporated into their personal schedule. The primary purpose of the action plan was to encourage participants to build new daily rituals related to changes in dietary habits (e.g., eating breakfast, more frequent eating times), exercise (e.g., scheduling at least three one-hour exercise periods per week), sleep (e.g., using new pre-sleep rituals), and recovery time (e.g., improving work-life balance) using a time management form. Each participant's fitness coach and RD helped participants follow their action plans related to exercise and dietary behaviors, respectively.

During the intervention, coaches monitored participant exercise activity and reviewed important DVM concepts from the program orientation each week. They also discussed ways that participants could make small changes in their physical activity habits and encouraged each individual through motivational statements and positive performance feedback. Participants were encouraged to follow their exercise prescription, including cardiovascular exercise at least three times per week and strength training a minimum of twice per week at a location of their choice. Coaches were present during the participants' regular workouts once per week at which time they provided additional exercise instruction and performance feedback.

At the start of the program, participants completed fitness pretests to serve as a baseline for post-intervention testing. Pre- and post-program fitness testing and weekly meetings based on the action plan were completed with fitness coaches at the Campus Recreation Center. The final meeting consisted of a second set of (post-intervention) fitness tests, ECI, and TSRQ measures.

Results

Exercise Commitment Index

A paired t-test assessed whether there was a significant difference in the participants' level of exercise commitment between pretest and posttest. The results showed a statistically significant increase in the participants' level of commitment from pretest to posttest, t(52) = -4.11, p < .001. Cohen's effect size value (d = .56) exceeds a moderate effect size level. The descriptive statistics for the pre- and post-intervention ECI data are presented in Table 1.

Reasons for Eating a Healthier Diet

All statistical analyses were performed using SPSS version 16.0 statistical software (SPSS Inc., Chicago, IL, USA). Because each of the major measures consisted of several dimensions, we conducted Multivariate Mixed Model (MMM) analysis with repeated measures on the time factor (i.e., pre- and post-intervention). For the TSRQ-Diet measure, the MMM showed that there was no significant time effect on the participants' perceived motivation levels related to the reasons for eating a healthier diet (Wilks' lambda likelihood ratio [LRATIO] = .972, p = .843, [[eta.sup.2] = .028). The descriptive statistics for the TSRQ-Diet measure are presented in Table 1.

Reasons for Exercising

The MMM analysis of the TSRQ-Exercise measure indicated a significant time effect on the participants' perceived motivation levels related to reasons for exercising (LRATIO = .805, p = .028, [[eta].sup.2] =. 195). The [[eta].sup.2] statistic indicated a large effect size (Cohen, 1988). The follow-up univariate tests showed that the amotivation measure contributed to the statistical significance, F(1, 52) = 10.48, p = .002 with observed [[eta].sup.2] = .168. Compared to pretest, participants reported that their posttest reasons for exercising were less likely to reflect low intentions or motivation. Other subscale scores did not change. The descriptive statistics for the TSRQ-Exercise measure are presented in Table 1.

Physical Fitness

The MMM analysis of the physical fitness measures indicated a significant time effect (LRATIO = .292, p < .001, [[eta].sup.2] = .708). The [[eta].sup.2] statistic indicated a large effect size. The follow- up univariate tests showed that the change from pretest to posttest was significant for all physical fitness measures: percent body fat, F(1,52) = 19.54, p < .001, [[eta].sup.2] = .273; VO2 max, F(1,52) = 16.61, p < .001, [[eta].sup.2] = .242; upper-body strength, F(1,52) = 54.72, p < .001, [[eta].sup.2] = .513; and lower-body strength, F(1,52) = 69.51, p < .001, [[eta].sup.2] = .572. The descriptive statistics for the physical fitness data are presented in Table 1.

Relationships among the Major Measures at Pretest and Posttest

The nature of the relationships between the ECI and other major measures also was examined at both pretest and posttest. As Table 2 shows, prior to the program, participants' level of exercise commitment was significantly related to diet and exercise autonomous motivation scores. Those with higher levels of exercise commitment reported more self-determined reasons for wanting to eat a healthy diet and exercise regularly. In addition, ECI scores and amotivation scores were significantly and negatively correlated for both the diet and exercise domains. That is, at posttest, increases in exercise commitment were associated with decreases in reporting a lack of intention and motivation with respect to eating a healthy diet or exercising regularly. None of the tests of the difference between the pretest and posttest correlation coefficients were significant.

Finally, to examine if participants who showed greater changes in fitness also showed significantly greater changes in exercise commitment and reasons for exercise, each of the four fitness variables were categorized into one of three groups: 0 = no improvement, 1 = small improvement, and 2 = large improvement. Participants in the no-improvement group either showed no change or showed worse fitness scores at posttest. The small-improvement participants showed positive change at posttest that was below the median of the entire sample, whereas large-improvement participants showed positive change at posttest that was at or above the sample median.

For the percent body fat measure, there were 17 no improvement, 6 small improvement, and 30 large improvement participants. For the V[O.sup.2] max measure, there were 9 no improvement, 17 small improvement, and 27 large improvement participants. For the upper body strength measure, there were 7 no improvement, 19 small improvement, and 27 large improvement participants. Finally, for lower body strength variable, there were 9 no improvement, 17 small improvement, and 27 large improvement participants.

An overall fitness change variable (FitChange) was created by summing the 0-2 scores from each of the four physical fitness variables, giving a possible score that ranged from 0-8 for each participant. Change scores in exercise commitment were computed by subtracting individuals' posttest scores on each item from their pretest scores. In addition, the difference in the average exercise commitment scores between pretest and posttest was computed.

Examination of the relationships between the FitChange and the individual exercise commitment items showed that as participants' fitness improvements increased, they reported at posttest a greater motivation to improve their health (r(51) = -.30, p = .03) and taking greater personal responsibility for their health, fitness, and well-being (r(51) = -.27, p = .05). The correlation between FitChange and the overall ECI change score was not significant, r(51) = -.22, p =.11.

With regard to participants' reasons for exercising, change scores were computed for each of the four subscales from the TSRQ-Exercise measure, by subtracting posttest scores from pretest scores. As participants' fitness scores (FitChange) improved, they reported significantly higher scores on the autonomous motivation subscale, r(51) = -.27, p = .05. None of the other subscales were significantly correlated with the FitChange measure.

Finally, the pre-program scores on the major measures were compared between the program completers and non-completers (n = 9). These comparisons showed only one significant difference--non-completers scored higher on the ECI (M = 4.37, SD = .74) than did the completers (M = 3.86, SD = .65), t(60) = 2.12, p = .04.

Discussion

Given the conceptual orientation of the DVM, participants in this program were expected to show improved fitness, as well as increased commitment and improved self-regulatory motivation. Results were generally consistent with expectations. After completing the program, participants reported an increased level of commitment and motivation to improve their health and fitness through regular exercise. More specifically, participants showed significantly improved fitness scores, increased commitment to exercising, and reductions in exercise amotivation. In addition, exercise commitment was positively associated with autonomous motivation and negatively associated with amotivation at posttest.

Contrary to expectations, however, participants did not show improved autonomous motivation after the program. The most likely explanation for this finding was a ceiling effect; the participants already showed high levels of autonomous motivation for both exercising and eating well at the start of the program. In addition, the final sample included only those individuals who completed the program. The comparison of the program completers with the non-completers showed that they differed only on their levels of exercise commitment. The fact that the small number of non-completers scored significantly higher than the completers on exercise commitment at pretest suggests that the former may not have needed the program as much as the latter.

The results also indicated that participants with larger changes in fitness showed greater exercise commitment and autonomous motivation to exercise at the end of the program. This is consistent with past research showing that affect and behavior are improved when individuals exercise for "internal" rather than "external" reasons (e.g., Ryan, Frederick, Lepes, Rubio, & Sheldon, 1997). On the other hand, researchers have found that non-exercisers show increased autonomous motivation after participation in physical activity programs, but much smaller changes in controlled motivation following program participation (Rodgers, Hall, Duncan, Pearson, & Milne, 2010). The current results showed minimal changes in both autonomous and controlled motivation after the program.

Amotivational reasons for improved eating habits and exercising regularly were expected to be reduced at the end of the wellness program. Prior to the program, participants were non-exercisers who likely did not value proper nutritional habits and regular exercise, and who lacked strong feelings of competence regarding their ability to change. In fact, participants showed increases in purposeful intentions and motivation regarding exercise after the program. In addition, increases in exercise commitment were negatively associated with amotivation for both eating and exercising at posttest. Although amotivation reasons for exercising were reduced at the end of the program, this reduction did not differ depending on the extent to which participants made significant fitness changes during the course of the program.

Taken together, the amotivation findings are consistent with previous research. For example, non-exercisers tend to focus on the disadvantages of exercising whereas regular exercisers tend to focus more on the advantages of exercising (Cropley, Ayers, & Nokes, 2003). Exercise amotivation can be thought of as a multidimensional construct, including outcome, capacity, effort and value beliefs (Vlachopoulos & Gigoudi, 2008). In future studies, researchers might examine what aspects of amotivation are most important in moving non-exercisers toward becoming regular exercisers in a values-based program. Previous research on non-exercisers participating in physical activity interventions has not included measures of amotivation (Rodgers et al., 2010).

The ECI used in the current study focused on the volitional "wanting to" rather than the obligational "having to" aspects of commitment. A previous measure of exercise commitment (Wilson et al., 2004) assessed both determinants and dimensions of commitment. Determinants included three items assessing wanting to commit (i.e., being determined, dedicated, or committed to exercising) and three items assessing having to commit (i.e., feeling obligated, necessary, or a duty to exercise). Because the latter components were not assessed, it remains to be determined whether duty or obligation aspects of commitment would increase or decrease as a result of participating in a DVM-based wellness intervention. Given that research has found that "wanting to commit" reasons were associated with increases in exercise behavior and "having to commit" reasons were not (Wilson et al., 2004), the latter form of commitment would be expected to decline after participation.

The results of the current study did not show any changes in reasons for eating a healthier diet. This was most likely due to the fact participants were required to meet only once with the program's registered dietician. The nutritional coaching was, therefore, less consistent and "weaker" than the exercise component of the program. Because no data on eating habit changes were collected from pretest to posttest, interpretation of the effects of participating in the program in this domain are limited.

Research on the quality of exercise coaching (e.g., Puente & Anshel, 2009) suggests that encouraging competence and autonomy among exercise participants is associated with positive affect and increased exercise frequency. In a study of primary care patients, both healthcare providers and physical activity counselors following a SDT program increased patients' autonomous motivation and subsequent physical activity levels (Fortier, Sweet, O'Sullivan, & Williams, 2007). Another study found that college students who took an exercise class with instructors who provided autonomy support had greater class participation, feelings of competence, and positive affect compared to a control class (Edmunds, Ntoumanis,& Duda, 2008).

The coaches who worked with the participants in the current program were trained to encourage self-determined regulation, consistent with both SDT and DVM frameworks (Deci & Ryan, 2008). However, participants' perceptions of autonomy and competence support from their coaches were not measured. To better account for the effects of the program coaches, future research using the DVM could measure coaching support, perhaps using the Sport-Exercise Climate Questionnaire (Baard, Deci, & Ryan, 2004).

Limitations of the Study

Because a randomized control group design was not used, there were limitations to the present study. In order to best test the effectiveness of the DVM, a control group would allow a clearer assessment of the effects of recognizing and acting upon one's disconnected values, receiving coaching, and formulating an action plan. Without such a comparison group, it is not possible to rule out that the mere passage of time or being motivated to participate in a wellness program accounted for the changes in commitment and reasons for exercising. Additionally, it is possible that merely participating in the pre-program fitness and other testing or completing the initial program orientation might have been at least partially responsible for the observed changes.

Designs such as a multiple baseline or a wait-list control were challenging due to the difficulty of requiring employees to avoid exercising during the intervention period. In addition, campus-wide announcements about an employee wellness program that provided extensive coaching at a very low cost made it very difficult to require individuals to register for a program and receive testing, yet have to wait until the following semester to engage in the program. Their incentive to experience the full program was immediate. Thus, conducting a study of this nature in such an enclosed environment as a university campus has inherent limitations.

When conducting action research, such as the current study, researchers develop a plan of action, implement the plan, observe the effects of action in the current context, and reflect on those effects as a basis for further planning and subsequent action (Herr & Anderson, 2005; Mills, 2003). Such research frequently must sacrifice research control and rigor (e.g., random assignment of coaches to participants, the systematic manipulation and assessment of single variables) in an effort to work within real-world settings and limitations, such as the closed environment of a university campus that encourages faculty interaction and reduces the likelihood of anonymity between treatment and control groups. Even more important was the fact that this university-supported program was well advertised and repeatedly promoted online throughout the campus community, making it impossible to offer a limited program that formed a control group; faculty participants wanted the full service. An action study was, therefore, appropriate and in accordance with the criteria of action research (Mills, 2003).

Despite the lack of traditional methodological rigor in action research studies, there are several forms of validation possible in this kind of research, including evaluative, outcome, and process validity. Each of these was evident in the present study. Evaluative validity, which addresses the objectivity of results, was supported in this study based on the quantitative nature of the data, in particular, significant improvements in fitness, commitment, and motivational measures. Outcome validity was obtained based on changes in specific behaviors that led to successful outcomes under study, and applicable to future research (e.g., significant improvement from pre-test to post-test). Process validity was apparent based on correctly conducting the intervention. Orientation content, pre- and post-test procedures, and fitness coaching were carried out properly.

Given the nature of the design, it is not possible to determine whether the observed increases in commitment and decreases in amotivation were the result of participating in the program, as opposed to stable changes in participants' commitment to and reasons for exercising. After the program ended, the participants' commitment could have fallen and their exercise reasons could have changed, negatively affecting their adherence.

It is also possible that the nominal participation fee ($25) might have increased participants' commitment and motivation to exercise. This fee was included because programs that require a minimal level of participant "buy-in" have greater adherence and lower drop-out rates (e.g., Chenoweth, 2002). However, given the amount of effort and commitment needed for participants and the size of the program fee, it is unlikely to have had a strong effect on participants' motivation or commitment.

Despite these limitations, it is reasonable to assume that the DVM program components made an important contribution to fitness improvements and changes in exercise commitment and reasons. A brief post-study program evaluation, submitted voluntarily by 37 of the 53 (70%) participants, indicated that 34 individuals rated the program as "satisfactory" or "very satisfactory," and that each of these individuals attempted to overcome a "disconnect" between their values and their unhealthy behavior patterns, as acknowledged during the orientation.

The DVM is a conceptual and intervention model that does not easily allow for testing its component parts. Nevertheless, the comprehensive nature of the program meant that multiple aspects of the intervention (e.g., orientation information, the action plan, the coaching, and so on) could have worked separately or in combination to bring about the observed changes in fitness and self-determined behavioral regulation. Future research will need to assess the relative importance of these different components.

In conclusion, the results provide additional insight into the effects of a values-based intervention derived from the DVM, and show that health behavior changes based on the DVM are associated with cognitive and psychological self-regulatory changes. The results support the argument that the DVM fosters positive health behavior change by increasing commitment to exercise as well as by reducing amotivational and capitalizing on autonomous reasons for wanting to exercise regularly. The current study adds to the emerging literature supporting the DVM (e.g., Anshel et al., 2010a; Anshel et al., 2010b; Brinthaupt et al., 2010). This literature now shows significant improvements in fitness levels, blood lipids, mental well-being, perceived barriers to exercise, and reasons for and commitment to exercising. Future research is needed to further confirm the extent to which the DVM, a values-based wellness intervention, will result in long-term changes and adherence in replacing unhealthy habits with a healthier, more desirable lifestyle.

Acknowledgement

This study was supported by a grant provided by Middle Tennessee State University to Sponsor an employee wellness program. The authors would like to express their thanks to Mr. Chris Dickson who served as the fitness coordinator and supervised the fitness coaching in this program and to Ms. Nan Allison, registered dietician, who provided the nutritional coaching and educational materials in the program.

References

ACSM's Guidelines for Exercise Testing and Prescription (2006). Philadelphia: Lippincott Williams & Wilkins.

Alexandris, K., Zahariadis, R, Tsorbatzoudis, C., & Grouios, G. (2002). Testing the sport commitment model in the context of exercise and fitness participation. Journal of Sport Behavior, 25, 217-230.

Allen, M. J., & Yen, W. M. (1979). Introduction to measurement theory. Long Grove, IL: Waveland.

Anshel, M. H. (2008). The Disconnected Values Model: Intervention strategies for health behavior change. Journal of Clinical Sport Psychology, 2, 357-380.

Anshel, M. H., Brinthaupt, T. M., & Kang, M. (2010). The Disconnected Values Model improves mental well-being and fitness in an employee wellness program. Behavioral Medicine, 36(4), 113-122. DOI: 10.1080/08964289.2010.489080

Anshel, M. H., & Kang, M. (2007). Effect of an intervention on replacing negative habits with positive routines for improving full engagement at work: A test of the Disconnected Values Model. Consulting Psychology Journal: Practice and Research, 59, 110-125.

Anshel, M. H., Kang, M., & Brinthaupt, T. M. (2010). A values-based approach for changing exercise and dietary habits: An action study. International Journal of Sport and Exercise Psychology, 8, 413-433.

Baard, P. P., Deci, E. L., & Ryan, R. M. (2004). Intrinsic need satisfaction: A motivational basis of performance and well-being in two work settings. Journal of Applied Social Psychology, 34, 2045-2068.

Brinthaupt, T. M., Kang, M., & Anshel, M. H. (2010). A delivery model for overcoming psycho-behavioral barriers to exercise. Psychology of Sport and Exercise, 11, 259-266.

Chenoweth, D. H. (2002). Evaluating worksite health promotion. Champaign, 1L: Human Kinetics.

Cohen. J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates.

Cropley, M., Ayers, S., & Nokes, L. (2003). People don't exercise because they can't think of reasons to exercise: An examination of causal reasoning within the transtheoretical model. Psychology Health and Medicine, 8, 409-414.

Deci, E. L., & Ryan, R. M. (2002). Handbook of self-determination research. Rochester, N-Y: University of Rochester Press.

Deci, E. L., & Ryan, R. M. (2008). Self-determination theory: A macrotheory of human motivation, development, and health. Canadian Psychology, 49, 182-185.

Dishman, R. K. (2001). The problem of exercise adherence: Fighting sloth in nations with market economics. QUEST, 53, 2

Ebbeling, C. B., Ward, A., Puleo, E. M., Widrick, J., & Rippe, J. M. (1991). Development of a single-stage treadmill walking test. Medicine and Science in Sport and Exercise, 23, 966-973.

Edmunds, J., Ntoumanis, N., & Duda, J. L. (2008). Testing a self-determination theory-based teaching style intervention in the exercise domain. European Journal of Social Psychology, 38, 375-388.

Fortier, M. S., Sweet, S. N., O'Sullivan, T. L., & Williams, G. C. (2007). A self-determination process model of physical activity adoption in the context of a randomized controlled trial. Psychology of Sport and Exercise, 8, 741-757.

Frederick, C. M. (2002). Self-determination theory and participation motivation research in the sport and exercise domain. In E. L. Deci & R. M. Ryan (Eds.), Handbook of self-determination research (pp. 277-296). Rochester, NY: University of Rochester Press.

Gagne, M., Ryan, R. M., & Bargmann, K. (2003). Autonomy support and need satisfaction in the motivation and well-being of gymnasts. Journal of Applied Sport Psychology, 15, 372-390.

Groppel, J. (2000). The corporate athlete: How to achieve maximal performance in business and life. New York: John Wiley & Sons.

Hagger, M., & Chatzisarantis, N. (2007a). Intrinsic motivation and self-determination in exercise and sport. Champaign, IL US: Human Kinetics.

Hagger, M., & Chatzisarantis, N. (2007b). Advances in self-determination theory research in sport and exercise. Psychology of Sport and Exercise, 8, 597-599.

Hall, P. A., & Fong, G. T. (2003). The effects of a brief time perspective intervention for increasing physical activity among young adults. Psychology and Health, 18, 685-706.

Herr, K., & Anderson, G.L. (2005). The action research dissertation: A guide for students and faculty. Thousand Oaks, CA.

Kelly, J., & Warnick, R. (1999). Recreation trends and markets: The 21s' century. Champaign, IL: Sagamore Publishing.

Levesque, C. S., Williams, G. C., Elliot D., Pickering, M. A., Bodenhamer, B., & Finley, P. J (2007). Validating the theoretical structure of the treatment self-regulation questionnaire (TSRQ) across three different health behaviors. Health Education Research, 21, 691-702.

Loehr, J., & Schwartz, T. (2003). The power of full engagement: Managing energy, not time, is the key to high performance and personal renewal. New York: Free Press.

Marcus, B. H., Ciccolo, J. T., Whitehead, D., King, T. K., & Bock, B. C. (2009). Adherence to physical activity recommendations and interventions. In S. A. Shumaker, J. K.

Ockene, & K.A. Riekert (Eds.), The handbook of health behavior change (3rd ed.; pp. 235-251). New York: Springer.

Miller, W. R., & Rollnick, S. (2002). Motivational interviewing: Preparing people for change (2nd ed.). New York: Guilford.

Mills, G. E. (2003). Action research: A guide for the teacher researcher (2nd ed.). Upper Saddle River, NJ: Merrill Prentice-Hall.

Mullen, E., & Markland, D. (1997). Variations in self-determination across the stages of change for exercise in adults. Motivation & Emotion, 21, 349-362.

Ockene, J. K. (2001). Strategies to increase adherence to treatment. In L. E. Burke & I. S. Ockene (Eds.), Compliance in healthcare and research (pp. 43-56). Armonk, NY: Futura.

Pollock, M. L., & Jackson, A. S. (1984). Research progress in validation of clinical methods of assessing body composition. Medicine and Science in Sports and Exercise, 16, 606-615.

Puente, R., & Anshel, M. H. (2009). Exercisers' perceptions of their fitness instructor's interacting style, perceived competence, and autonomy as a function of self-determined regulation to exercise, enjoyment, affect, and exercise frequency. Scandinavian Journal of Psychology, 51, 38-45.

Rhodes, R., & Matheson, D. (2005). Discrepancies in exercise intention and expectation: Theoretical and applied issues. Psychology & Health, 20, 63-78.

Rodgers, W. M., Hall, C. R., Duncan, L. R., Pearson, E., & Milne, M. I. (2010). Becoming a regular exerciser: Examining change in behavioural regulations among exercise initiates. Psychology of Sport and Exercise, 11, 378-386.

Rollnick S., Mason P., & Butler, C. (1999). Health behavior change: A guide for practitioners. Edinburgh: Churchill Livingstone.

Ryan, R. M., Frederick, C. M., Lepes, D., Rubio, N., & Sheldon, K. M. (1997). Intrinsic motivation and exercise adherence. International Journal of Sport Psychology, 28, 335-354.

Ryan, R. M., Williams, G. C., Patrick, H., & Deci, E. L. (2009). Self-determination theory and physical activity: The dynamics of motivation in development and wellness. Hellenic Journal of Psychology, 6, 107-124.

Scanlan, T. K., Simons, J. P., Carpenter, P. J., Schmidt, S. W., & Keeler, B. (1993). The Sport Commitment Model: Measurement development for the youth-sport domain. Journal of Sport & Exercise Psychology, 15, 16-38.

Self-Determination Theory (n.d.). The Self-Regulation Questionnaires. Available at: http://www.psych.rochester.edu/SDT/measures/SRQ_healthTSRQ.php. Accessed on August 30, 2010.

Vlachopoulos, S., & Gigoudi, M. (2008). Why don't you exercise? Development of the Amotivation Toward Exercise Scale among older inactive individuals. Journal of Aging and Physical Activity, 16, 316-341.

Wilson, R M., Mack, D. E., & Grattan, K. P. (2008). Understanding motivation for exercise: A self-determination theory perspective. Canadian Psychology, 49, 250-256.

Wilson, E M., Rodgers, W. M., Carpenter, P. J., Hall, C., Hardy, J., & Fraser, S. N. (2004). The relationship between commitment and exercise behavior. Psychology of Sport and Exercise, 5, 405-421.

Winett, R. A., Anderson, E. S., Wojcik, J. R., Winett, S. G., & Bowden, T. (2007). Guide to health: Nutrition and physical activity outcomes of a group-randomized trial of an internet-based intervention in churches. Annals of Behavioral Medicine, 33, 251-261.

Thomas M. Brinthaupt

Minsoo Kang

Mark H. Anshel

Middle Tennessee State University

Address Correspondence to: Thomas M. Brinthaupt, Department of Psychology, P.O. Box X034, Middle Tennessee State University, Murfreesboro, Tennessee, USA 37132. Phone: 615.898.2317; Fax: 615.898.5027; tom.brinthaupt@mtsu.edu

Table 1
Descriptive Statistics for Level of Exercise Commitment, Reasons
for Eating a Healthier Diet, Reasons for Exercising, and Physical
Fitness (N = 53)

                                Pretest           Posttest

Variable                       M        SD       M        SD

Exercise Commitment Index     3.86     0.65     4.27     0.47

TSRQ-Diet
  Autonomous motivation       6.07     0.90     5.97     1.09
  Introjected regulation      4.07     1.74     4.08     1.65
  External regulation         2.65     1.38     2.57     1.34
  Amotivation                 2.09     1.07     2.06     1.00

TSRQ-Exercise
  Autonomous motivation       6.14     0.88     6.15     0.92
  Introjected regulation      4.07     1.55     4.20     1.36
  External regulation         2.52     1.26     2.47     1.32
  Amotivation                 2.18     1.11     1.70     0.78

Physical Fitness
  Percent body fat           29.02     7.81    27.31     8.13
  V[O.sup.2] max             35.78     9.54    38.97     9.20
  Upper-body strength        56.64    22.89    72.35    27.94
  Lower-body strength       191.69   100.11   268.87   120.57

Note. TSRQ = Treatment Self-Regulation Questionnaire; higher scores
on the Exercise Commitment Index indicated a larger commitment;
higher scores on the reasons for eating a healthier diet and for
exercising indicated a higher motivation level.

Table 2
Pretest and Posttest Correlations of Exercise Commitment
Index Scores with Diet, Exercise, and Fitness Measures

                                  Exercise
                              Commitment Index

Variable                     Pretest   Posttest

TSRQ-Diet
  Autonomous motivation       .39 **     .37 **
  Introjected regulation     0.14       0.03
  External regulation       -0.03      -0.10
  Amotivation               -0.15      -0.27

TSRQ-Exercise
  Autonomous motivation       .28 *      .41 **
  Introjected regulation     0.10       0.15
  External regulation       -0.09      -0.09
  Amotivation               -0.16      -0.35

Physical Fitness
  Percent body fat          -0.06       0.13
  V[O.sup.2] max            -0.08      -0.11
  Upper-body strength       -0.06      -0.02
  Lower-body strength       -.01       -0.07

Note. N = 53; TSRQ = Treatment Self-Regulation Questionnaire;
higher scores on the Exercise Commitment Index indicated a
larger commitment; higher scores on the reasons for eating
a healthier diet and for exercising indicated a higher
motivation level; * p < .05; ** p < .01.