Printer Friendly

Varying the Mode of Cardiovascular Exercise to Increase Adherence.

Many people do not engage in the recommended frequency, intensity and duration of exercise required to produce the psychological and physiological benefits of an exercise regimen. The purpose of this study was to investigate whether varying workouts over an 8-week period would increase adherence levels in sedentary individuals. Three different modes of exercise were examined: variable, static, and preferred Participants in the variable condition performed the same exercise for 2 weeks, at which time they changed exercises. Those in the static condition chose an exercise and remained with that single exercise for the duration of the study Participants in the preferred mode condition were not given a protocol as to when to change their exercise and were free to determine their mode of exercise. In addition to two measures of adherence, variables such as cardiovascular fitness, enjoyment, boredom, and physical self-efficacy were examined. Results indicated that the variable aerobic exercise regimen increased adh erence to the workout program as compared to the preferred mode. Furthermore, the participants in the variable program rated their workouts as more enjoyable. In addition, cardiovascular fitness increased across conditions, suggesting that all groups improved above baseline values. Boredom did not impact adherence, nor was boredom related to exercise program variability. No differences were found in physical self-efficacy over time, suggesting that it may remain stable over relatively short durations. Future research directions and implications are discussed with respect to the potential benefits of employing a variable exercise program.

Habitual physical exercise has long been praised for its positive impact on quality of life. The benefits of exercise range from psychological to physiological health conditions. For instance, researchers have suggested that regular exercise can reduce anxiety (Petruzzello, Landers, Hatfield, Kubitz, & Salazar, 1991), relieve both acute and chronic symptoms of depression (Camacho, Roberts, Lazarus, Kaplan, & Cohen, 1990), and increase self-esteem and physical self-concept (Spence, Poon, & Dyck, 1997). Additionally, it can control obesity (Brownell & Stunkard, 1980), hypertension (Boyer & Kasch, 1970), and diabetes (Vranic & Berger, 1979). Exercise can also be used in the prevention of cardiovascular disease (Oberman, 1985) and can decrease mortality and morbidity rates (Blair, Kohl, Paffenbarger, Clark, & Cooper, 1989). This aggregate body of research substantiates the importance of engaging in a habitual exercise regimen.

In spite of the myriad of health benefits of exercise, adult participation rates are disappointing. Recent reports indicate that greater than 60% of adults do not engage in the recommended amount of physical activity and 25% are completely sedentary. Furthermore, only 15% of adults engage in regular, vigorous exercise during leisure time (Centers for Disease Control, 1996). In addition, researchers have determined that approximately 50% of participants who begin an exercise program drop out within the first 3 to 6 months (Dishman, 1988). The severity and prevalence of sedentary lifestyles and non-compliance rates have motivated researchers to further investigate factors that impact exercise adherence.

Much of the research conducted in the area of exercise adherence can be classified into two categories: determinants and interventions. Determinants include factors such as demographics, personality characteristics, and barriers. For example, socioeconomic status, gender, age, race, and exercise history correlate with participation rates (e.g., Dishman & Buckworth, 1997; Welsh, Labbe, & Delaney, 1991). Characteristics of the individual may play a particularly large role in habitual exercise, as self-efficacy appears to be one of the most frequently identified psychosocial determinants of exercise adherence. Self-efficacy has been defined as the conviction that one holds in her/his ability to successfully execute the desired behavior required to produce a certain outcome (Bandura, 1997). This appears to be important in both the adoption (McAuley, 1992) and the maintenance (McAuley, Lox, & Duncan, 1993) of habitual exercise. McAuley, Coumeya, Rudolph, and Lox (1994) suggest that by implementing a selfefficacy intervention, exercise frequency increases. McAuley (1992) maintains that self-efficacy is a significant predictor of initial exercise adoption. Additionally, Rodgers and Gauvin (1998) revealed that self-efficacy levels discriminated moderately active from highly active participants. The importance of self-efficacy should not be underestimated when designing adherence interventions.

Enjoyment may also be a key in increasing exercise adherence rates among sedentary individuals. While it appears that many researchers tend to agree that enjoyment is a necessary ingredient to establishing a lifestyle of fitness, little research has been conducted in the area (e.g., Wankel, 1993). In fact, much of the existing research in exercise enjoyment has focused on the definition and measures of enjoyment instead of its relationship with exercise (e.g., Crocker, Bouffard, & Gessaroli, 1995; Kendzierski & Decarlo, 1991; Kimiecik & Harris, 1996; Wankel, 1997). However, evidence exists that those who enjoy exercise will continue with it as compared to those who do not enjoy exercise (Morgan, Shephard, Finucane, Schimmelfing, & Jazmaji, 1984) and that enjoyment motives are important for the maintenance of the activity (Ingledew, Markland, & Medley, 1998). At the opposite end of the spectrum, boredom is a frequently cited barrier not only to regular exercise (King, Taylor, Haskell, & Debusk, 1990), but als o to specific forms of exercise such as jogging (Vitulli & DePace, 1992). While the research in this area is scarce, it is becoming increasingly evident that finding ways of maximizing enjoyment while minimizing boredom may prove to increase exercise participation among sedentary individuals.

In addition to the abundance of research conducted on the determinants of habitual exercise, investigators also focus on interventions that may increase participation rates. For example, strategies such as telephone reminders (e.g., Lombard, Lombard, & Winett, 1995), relapse prevention (e.g., Marcus & Stanton, 1993; Simkin & Gross, 1994), and goal setting (e.g., Gallucci, 1995; Martinet al., 1984; Perkins, Rapp, Carlson, & Wallace, 1986) appear to have a positive affect on exercise adherence. Other interventions that have been successful at increasing exercise adherence rates include incentives and contracts (e.g., Neale, Singleton, Dupuis, & Hess, 1990; Robison et al., 1992), self-monitoring (e.g., Weber & Wertheim, 1989), and home-based versus group-based programs (e.g., King et al., 1997, King, Haskell, Taylor, Kraemer, & DeBusk, 1991; King, Haskell, Young, Oka, & Stefanick, 1995; Perri, Martin, Leermakers, Sears, & Notelovitz, 1997). While this list is not exhaustive, it does indicate the abundance of re search designed to increase adherence levels.

These types of interventions have had a positive impact on exercise adherence, but many contain limitations. For instance, several interventions require either supervision or participation in a group, yet research strongly suggests that those engaged in exercise prefer to do so at home as compared to a formal group setting (King et al., 1997). Techniques such as relapse prevention, reinforcement, incentives, and fitness testing require the participant to attend meetings or exercise in a supervised setting. These types of interventions may favor only those who can afford to participate, obtain a membership to a gym, or work in a setting which offers access to a fitness facility. While other techniques such as telephone contact or mediated approaches (mailings, signs) allow the participant to exercise at home, they require the involvement of an outside party (Cardinal & Sachs, 1995, 1996). It may not be practical or realistic to offer these options to every adult who wishes to participate. Also, interventions such as these cannot continue indefinitely. Participants should be offered interventions on which they can rely for unlimited support.

In light of the flaws in practicality of the interventions discussed above, participants need simple, effective, practical, flexible, and unbiased guidelines to help them adhere to exercise regimens. There is a need to focus on interventions that participants can conduct at home, without supervision, that can occur regardless of location or socioeconomic status, and can continue for as long as the participant wishes. One such intervention that could possibly accommodate these qualifications is variability in the mode of exercise. For the purpose of this study, variability can be defined as a periodic change in an exercise regimen. For sedentary individuals who find boredom as an inhibitor to habitual exercise, a periodic change in a fitness program might help maintain interest. By introducing a dynamic exercise regimen that continually changes, or varies, adherence rates might increase. A variable program is one that can be implemented by the participant and can continue throughout life, thereby encouraging lifetime fitness. Variability does not require the supervision of health professionals, a membership to a gym, or a worksite with available fitness facilities. Likewise, variable programs can be effective regardless of the location of exercise, socioeconomic status, race, or gender. Given these potential benefits, this investigation was designed as an initial examination of the impact of variability on an exercise regimen.

The specific purpose was to determine whether introducing variability to an aerobic exercise regimen (variable) would increase adherence in sedentary adults as compared to those who routinely perform the same exercise (static) or those who were not given a protocol for when to vary their workout (preferred). While one can develop many different operational definitions of variability, for the purposes of this study variability was operationally defined as a change in cardiovascular exercise every two weeks. Also investigated was the impact of a variable program on self-efficacy, boredom, enjoyment, and cardiovascular fitness. It was hypothesized that those in the variable group would exhibit greater exercise adherence than would groups that used a static or preferred mode. Furthermore, those in the variable condition were predicted to rate their exercises as more enjoyable and less boring, to have a greater increase in cardiovascular fitness, and to have a higher level of self-efficacy, as compared to the sta tic and preferred conditions.



Initially, 113 men and women between the ages of 18 and 35 were selected for this study (37 in the variable, 38 in the static, and 38 in the preferred mode). Participants were recruited from university fitness centers via signs posted on facility doors and information tables set up inside the fitness buildings. Interested volunteers completed a health history questionnaire developed specifically for the study that assessed age, past exercise history, and potential health risks that may interfere with regular vigorous exercise. Volunteers who were university students between the ages of 18 to 35, had no health risks that would interfere with an exercise routine, and were sedentary, were included in the investigation. For the purposes of this study, individuals were classified as sedentary if they had not engaged in any type of aerobic, resistance, or sport activity more than once a week for the previous three months. Participants with activity levels exceeding once a week for the previous three months were co nsidered active and not included in the study. Those who met the age requirements, had no health risks, and had been sedentary for the previous three months were scheduled for an initial meeting I.

It should be emphasized that although the term "sedentary" was used to describe the relative fitness characteristics of the sample, some of these individuals may have engaged in occasional sporadic exercise. The operational definition of "sedentary" as it is used in the current investigation, is a relative term that does not strictly specify a complete lack of physical activity.

At the conclusion of the investigation, 62 participants including 13 men and 49 women with a mean age of 21.13 (SD = 3.38) remained active in the study. Prior to analysis, 24 participants were in the variable, 22 in the static, and 16 in the preferred mode conditions. However, one participant in the preferred mode condition was removed from the study because her total adherence score was greater than two standard deviations from the mean preferred mode adherence score. Therefore, 61 participants were included in the final analysis (24 in the variable, 22 in the static, and 15 in the preferred mode condition). Refer to Table 1 for the number of men and women and ages for each condition. These figures constituted a 35% dropout rate for the variable condition, a 42% dropout rate for the static condition, and a 58% dropout rate for the preferred mode condition. Overall dropout rates were equal to 45% which is comparable to the 50% dropout rate cited in the literature (Dishman, 1988).

It is important to note that one participant in the static condition did not complete the posttest measure of cardiovascular fitness due to a back injury the previous day. In addition, one participant in the preferred mode condition had a total adherence score of zero. Because she did not exercise one time, she did not have an enjoyment or boredom score. While these two participants were included in the overall analysis, their data for posttest step test, enjoyment, and boredom were incomplete, and consequently, not included in the analysis.

Instruments and Tests

Adherence. The primary dependent variable of interest was adherence. Two different operational definitions of adherence were used for this study (Martin et al., 1984). Ideal adherence was calculated as the total number of weeks that participants exercised at least three times. A second operational definition of adherence, total adherence, was calculated as the total number of sessions attended over the duration of the study. This measure is more sensitive than the first because it accounts for those who exercised more than three times a week.

An exercise log card was developed for this study in order to record adherence data. It was a self-report log on which participants entered each completed exercise bout. They were asked to record the date, time, exercise mode, duration, a 10-second exercise heart rate, rating of perceived exertion (RPE) during the workout (Borg, 1985), and enjoyability and boredom rankings. The log card remained at the facility and was completed by the participant immediately following each bout of exercise.

Participants were instructed on how to complete each log card and were shown how to determine RPE and how to read the chart. Furthermore, they were shown how to find their target heart rate zone and were instructed to take a 10-second heart rate and determine their RPE in the middle of their workout.

Enjoyment and Boredom. There were two items on the exercise log card that assessed enjoyment and boredom levels. At the completion of each bout of exercise, participants were asked to determine how bored they were with the exercise and how much they enjoyed the exercise. These two items were adopted from the Physical Activity Enjoyment Scale (PACES; Kendzierski & DeCarlo, 1991). To examine enjoyment levels, a seven-point Likert scale was adopted from the PACES, ranging from one (I enjoy it) to seven (I hate it). A similar seven-point Likert scale was adopted for the boredom construct, which ranged from one (1 feel bored) to seven (I feel interested). Because participants completed the questions every time they exercised, only one item representing each construct was used. This prevented a practice effect from occurring.

Self-Efficacy. To determine self-efficacy differences among participants, the Physical Self-Efficacy Scale (PSE) was used (Ryckman, Robbins, Thornton, & Cantrell, 1982). This scale is a general measure of self-efficacy and includes two subscales: the Perceived Physical-Ability subscale and the Physical Self-Presentation Confidence subscale. The PSE is a 22-item Likert scale and is a well-established measure of efficacy that has been used in numerous studies in the past (e.g., McAuley, 1992). The test was administered at the beginning and again at the completion of the 8-week study.

Cardiovascular Fitness. Cardiovascular fitness was estimated using participants' recovery heart rate. The test chosen was the submax YMCA 3-Minute Step Test (Golding, Myers, & Sinning, 1989), which requires participants to step onto a 12-inch high platform for 3 minutes at a rate of 24 steps per minute. Participants are instructed to sit immediately after the three minutes. Within five seconds of the completion of the test, a skin palpitated recovery heart rate was taken at the radial artery for one minute. The 3-minute step test was administered both at the onset of the study and again at its completion, and the changes in recovery heart rate were used to estimate changes in cardiovascular fitness levels. Procedure

Participant recruitment began during the first week of the semester. Information tables were set up outside of the fitness facility promoting participation in this study. Patrons who stopped by the tables and expressed an interest in the study were directed to complete a health history questionnaire. Volunteers who had been sedentary for the previous three months, had no health risks, and were between the ages of 18 and 35 were given an appointment to meet with the investigator.

During this meeting, participants completed the informed consent, the PSE, and the 3-minute step test. Each participant then was instructed to exercise at either one of two fitness facilities on campus. Within these facilities, participants had their choice of treadmills, stairmasters, stepmills, stationary bicycles, row ergometers, and various aerobic classes. Participants were instructed to exercise 3 times a week, 20 to 60 minutes each session at 60% to 90% of their maximum heart rate (Pollock et al., 1998). Participants were shown how to determine if they were exercising at 60% to 90% of their maximum heart rate by taking a carotid or radial heart rate for 10 seconds. They were also given exercise mode guidelines according to their respective conditions, and were shown how to complete the exercise log card. Finally, they were reminded that they would be contacted in eight weeks to repeat these procedures.

Participants were randomly assigned to one of three treatment conditions: variable mode, static mode, or preferred mode. In the variable mode condition, participants were asked to engage in one cardiovascular exercise for two weeks. At the conclusion of those two weeks, they were instructed to select another new mode of cardiovascular exercise. In repeating this cycle, participants could select a minimum of two different exercises and a maximum of four different exercises. For example, they could have merely alternated between the stationary bicycle and treadmill, or they could have chosen a different exercise every two weeks (i.e., stationary bicycle, stairmaster, stepmill, and row ergometer). In the static mode condition, participants were asked to select a mode of cardiovascular exercise and remain with that single exercise for the duration of the study. Finally, the preferred mode condition served as a control group. They were given the identical protocol concerning frequency, intensity, and duration com pared to the other conditions, but they were not given instructions as to whether to or when to vary their workouts. Participants could chose whatever mode of exercise they preferred each time they visited the facility, as long as it was one of the approved exercises. Participants in all conditions received the same amount of interaction with the investigators, and no special attention was given to explain the more complex variable condition. All participants received a sheet of paper outlining the requirements for their respective condition, in addition to information regarding target heart rate, exercise frequency and duration, and RPE.

During the following eight weeks, participants exercised in the facility at times, intensities, and durations of their choice, all of which were recorded on the log card. The exercise activities participants could choose from included the treadmill, stairmaster, stepmill, stationary bicycle, row ergometer, and aerobic classes. Log cards were completed at the end of each bout of exercise. At the end of the testing session, the principle investigator contacted participants and a final meeting was established. During the last meeting, participants completed the PSE and the 3-minute step test. At the conclusion of the final meeting, the nature of the study was discussed with participants, including all manipulations. Participants were told to contact the principle investigator with questions about the results or any future concerns.


Please see Table 2 for means and standard deviations associated with each of the different dependent measures across the three groups of interest.


Two measures of adherence were operationalized: ideal and total adherence. To reiterate, ideal adherence was defined as the number of weeks participants exercised three or more times. Total adherence was defined as the total number of exercise sessions participants completed. Due to strong a priori predictions regarding adherence, and in accordance with the recommendations of Myers and Well (1995), a series of planned comparisons was conducted on the ideal adherence measure. The planned comparisons revealed no significant differences in ideal adherence between the variable and static conditions, t (44) = .76, p[greater than].05, and the variable and preferred mode conditions, t (37) = 1.44, p[greater than] .05. Bonferroni's procedure for alpha adjustments was conducted to control for potential inflation of the Type I error probability.

A similar series of a priori planned comparisons were conducted on the total adherence measure. Analysis revealed that although participants in the variable condition did not exercise significantly more than participants in the static condition, t (44) = .81, p [greater than] .05, they did exercise more than participants in the preferred mode condition t (37) 2.05, p [less than] .025. Again, adjustment of alpha levels was made to account for multiple analyses, as per Bonferroni's guidelines. The analysis suggests that those in the variable condition maintained a greater exercise frequency than those in the preferred mode condition (Please see Figure 1).


To examine differences in enjoyment of the exercise regimen among participants in the three conditions, a one-way ANOVA was conducted on the enjoyment measure. Note that a lower enjoyment score corresponds to a higher degree of enjoyment. Results revealed significant differences in enjoyment among conditions, F (2, 59) = 4.48, p [less than] .05. Post hoc comparisons using Tukey's HSD procedure revealed that participants in the variable condition enjoyed their workouts more than those in the preferred mode condition (Please see Figure 2).

While there was only a single item used to test enjoyment and boredom, it is important to note that these items were significantly negatively correlated, r= .725, p[less than].01.


To determine whether the exercise protocols led to differences in boredom with the exercise regimen, a one-way analysis of variance (ANOVA) was conducted on the boredom measure. The analysis revealed no significant differences in boredom among the participants in the three conditions, F(2, 59) = l.33, p[greater than] .05. This suggests that participants reported equal levels of boredom across the three conditions. However, it is important to note that overall, participants appeared interested in their protocol (GM= 5.29, SD = .99 on a scale of 1 = "I feel bored" to "7 = 1 feel interested").

While there was only a single item used to test enjoyment and boredom, it is important to note that these items were significantly negatively correlated (r = -.725, p [less than] .01).

Physical Self-Efficacy

It was hypothesized that physical self-efficacy would be a significant predictor of adherence. To test whether this was the case, a simple regression analysis was run using the physical self-efficacy pretest as the predictor variable and adherence as the criterion variable. Analysis revealed that physical self-efficacy did not significantly predict total adherence, [R.sup.2] = .04, F(1,60) = 2.63, p[greater than] .05.

To determine whether different exercise regiments affected physical self-efficacy over time, a 3 X 2 (Condition X Time) mixed model ANOVA with repeated measures on the last factor was conducted using pre and post measures of physical self-efficacy as the dependent measure. Neither the Condition, F(2,58) = .63, p[greater than] .05 nor the Time main effect, F(1, 58) = .75, p[greater than] .05 were significant. Furthermore, there was not a significant interaction, F(2,58) = .46, p[greater than] .05. This suggests that physical self-efficacy did not change significantly over time.

Cardiovascular Fitness

Because past research has shown that increased exercise enhances cardiovascular fitness (Powers & Howley, 1997), an analysis was conducted to determine if there was a significant condition and time interaction. Participants in the variable group exercised significantly more than those in the preferred mode condition. Therefore, they should have exhibited a reduction in recovery heart rate (greater increase in cardiovascular fitness). To determine if this was the case, a 3 X 2 (Condition X Time) mixed model ANOVA with repeated measures on the second factor was conducted on the 3-minute step test. The analysis revealed a main effect for Time, F(1, 57)=22.29, p [less than] .001, suggesting that recovery heart rate levels improved for all groups over the duration of the study. However, there was not a significant main effect for Condition, F(2, 57)=.80, p [greater than] .05 nor was there a significant interaction F(2,57)=.1 3, p [greater than] .05. These findings suggest that all participants improved in cardiov ascular fitness over the duration of the study.

To ensure that all participants were exercising at a similar intensity level, a one-way ANOVA was conducted on the RPE. Results revealed no significant differences among conditions, F(2,59) = .007, p [greater than] .05, suggesting that participants in the variable, static, and preferred mode were exercising at the same perceived intensity level. Participants exercised at an RPE of M = 14.42, SD = 1.62, which approximates "hard" on the RPE scale.

Manipulation Check

A manipulation check was devised to determine how closely participants followed their exercise regiment. The manipulation check was based on the number of different exercises performed by each participant over the duration of the study. For example, those in the variable mode should have completed between two and four different exercises, those in the static should have engaged in only one exercise, and those in the preferred mode could have taken part in any number of different exercises. A one-way ANOVA conducted on the means for each condition revealed no significant group differences, F(2, 59) = .687, p [greater than] .05, suggesting that the static group did not strictly maintain its single exercise guideline, and also indicating that individuals may inherently desire variability in their exercise program.


The present investigation was designed to determine whether introducing variability into an aerobic exercise regimen would increase adherence over an eight-week study as compared to a static or preferred mode group. Variability was operationalized as changing an aerobic exercise routine every two weeks. This was in contrast to a static group that remained with the same exercise for the duration of the study, and a preferred mode group that was permitted to choose what exercise they preferred each visit to the facility. Findings supported the hypothesis that the variable condition would demonstrate greater total adherence as compared to the preferred mode group. Contradicting the hypothesis, the variable condition did not have greater total adherence than the static group. Furthermore, analyses revealed that those in the variable group enjoyed their workouts more than those in the preferred mode group. There was a significant increase in cardiovascular fitness over time, but no interaction among conditions. F inally, no significant differences in boredom were revealed, and physical self-efficacy failed to predict adherence.

These findings provide several avenues for discussion. To begin, the most pertinent findings with regard to adherence levels will be addressed. Increases in adherence may be attributed to a more structured workout regimen. For example, those in the variable condition were given a strict structure as to when to change their exercises. In a sense, they were given short-term goals to achieve; complete the next two weeks and then change the exercise. This, in effect, appeared to increase the total number of times they exercised over the eight-week period as compared to those in the preferred mode condition. Members of the preferred mode group were not given a protocol to help them structure their exercise routine. Therefore, they had no short-term goals to accomplish. Apparently, this open-ended structure induced an environment in which they did not exercise as many times compared to the variable condition.

Increases in adherence may also be resultant of the perception of a more structured workout regimen. Results indicated that the manipulation was unsuccessful for the static condition and that all conditions chose approximately the same number of exercises over the duration of the study. Even though the number of exercises was the same, adherence was increased for the variable condition. In this case, it may not be the actual structural manipulation that caused the increase in adherence, but the perception of structure. More research needs to be conducted to determine whether it is the structure or the perception of structure causing increases in adherence.

Yet another viable option for increases in adherence may be successful goal attainment. Previous research has shown that goal setting increases exercise adherence (e.g., Perkins et al., 1986), and with those in the variable condition setting goals to change exercises, this may have affected the adherence levels. It is also important to note that there were differences in the number of men and women in each condition. The preferred condition had a disproportionately higher ratio of men to women as compared to the static and preferred mode conditions, which may have subtly influenced adherence levels.

The need to have a more structured fitness program may be attributed to the population examined. Because the participants were sedentary, it is possible that they needed a more structured regimen to follow in order to establish exercise in their daily routine. However, this notion can only be postulated at this point, as the exercise behavior of regular exercisers was not evaluated in the current investigation. Research efforts that include regular exercisers who already have an established structure and schedule could determine whether or not structure is an underlying cause of adherence differences.

It is interesting to note that while total adherence increased as a function of condition, there were no statistical differences in ideal adherence. While the concepts of total and ideal adherence were adopted from Martin et al. (1984), it would be inappropriate to compare the present findings to their study because their exercise regimen used a completely different protocol. For example, the present study entrusted participants to exercise three times a week without supervision while Martin et al. (1984) asked participants to attend an exercise class twice a week and exercise without supervision only once a week. Because there may be inherent differences in ideal adherence between supervised exercise and unsupervised exercise, it is obviously quite difficult to make logical comparisons and contrasts.

A reason that total adherence increased in the variable group may be attributed to the increase in enjoyment they experienced. It was hypothesized that those in the variable condition would enjoy their exercises to a greater extent as compared to those in the other two conditions. Analyses suggested this to be the case. Wankel (1993) indicated that those who enjoy their exercises have higher adherence levels. This study appears to support this notion, but the precise cause for the increase in enjoyment only can be speculated. The increase in enjoyment may be attributed to a constant change in exercise. Another possibility to consider is that a more structured regimen (i.e., variability is prescribed and how to vary exercise modes is structured) may increase enjoyment. It is also possible that goal attainment may have had a positive impact on enjoyment. Finally, higher enjoyment rankings in the variable group as compared to the preferred mode group may be attributed to the success of exercising regularly. Eac h of these possibilities warrants further investigation.

It was also hypothesized that boredom levels would decrease in the variable condition as compared to the other two conditions. The analyses did not support this hypothesis. All participants in this study cited a statistically similar boredom level for their exercises. Yet these findings appear counterintuitive. Changing the exercise routine increased enjoyment but did not decrease boredom, despite a significant negative correlation between the two constructs. While boredom was similar across groups, enjoyment differentiated them. In spite of the fact that boredom is one of the more frequently cited reasons for dropping out of exercise programs (King et al., 1990), it appears that it did not have a meaningful effect on adherence in the current investigation. An interpretation of this finding may be that boredom is not a determining factor when examining adherence, suggesting that while enjoyment and boredom are significantly correlated, they are mutually exclusive constructs.

It was predicted that cardiovascular fitness would increase in the variable condition as compared to the static and preferred mode conditions. The analyses revealed that indeed there was an increase in cardiovascular exercise over time. These findings were expected because all of the groups exercised more during the program than before testing. Yet, no statistical difference was revealed for an interaction among conditions. This may be attributed to the sensitivity of the step test measure. Because the 3-minute step test is a submaximal measure of fitness levels, there is standard deviation of approximately 11 beats per minute error associated with the tests (Londeree & Moeschberger, 1984).

A final hypothesis was that those with high physical self-efficacy would demonstrate greater adherence when compared to those with low physical self-efficacy. The analysis did not support this hypothesis, which is inconsistent with the majority of previous literature (McAuley, 1992). The failure to achieve significance may be attributed to the length of the study. Previous investigations examining physical self-efficacy and adherence have lasted as long as five months (e.g., McAuley, 1992; McAuley et al., 1994). Given that the current investigation lasted eight weeks, it is difficult and inappropriate to make direct comparisons. In support of the current results, McAuley & Jacobson (1991) found that there were no significant differences on pre-program self-efficacy measures between good and poor adherers for an 8-week exercise program. It is interesting to note that in the present study, physical self-efficacy did not change from the pretest to the posttest measure. This suggests that physical self-efficacy as measured by the PSE, may be a stable construct, resisting change over relatively short periods of time.

Findings from the current investigation warrant one further comment. While uneven sample sizes prove problematic in data analysis, it is important to speculate why there was such a large dropout rate in the preferred mode condition as compared to the static or variable conditions. Across all conditions, every attempt was made to encourage participants to return for their final meeting with the investigator. Those in the preferred mode condition appeared particularly resistant to set up meetings, return phone calls, or even attend meetings already established. It is doubtful that this was because of initial differences in conditions because participants were randomly assigned to the groups. Furthermore, no incentives were offered to any condition to encourage participants to complete the study, so differences cannot be attributed to incentives. Whether the failure to attend the final meeting was due to guilt for not exercising during the study, or embarrassment for not meeting the experimenter's expectations of performance, it should be investigated why there was such a large difference in dropout rates.

Results suggest that a variable program in which the exercise mode is changed every two weeks will increase adherence as compared to an unstructured program in which individuals self-select the particular mode of exercise in which to engage. This difference may be attributed to a structured regimen, the perception of a more structured regimen, or it may be attributed to the increase in enjoyment that arises from a variable exercise regimen. However, it appears that a variable program does not affect how bored participants are during their workouts, nor does it result in short-term changes in physical self-efficacy. Finally, cardiovascular fitness increased over time across all conditions, suggesting that all groups improved in fitness levels.

Several directions for future research in the area of exercise adherence, with particular relevance to the structure of exercise schedules, can be speculated based on the results of this initial investigation. For instance, it would be important to determine whether it is a continuous change in exercise, the nature of a structured workout regimen, or the perception of a structured workout regimen that maintains participants' interest and therefore increases adherence. Why enjoyment increased while boredom did not decrease still remains a question, even though these two items were correlated. Also, it would be valuable to investigate the enjoyment and boredom levels of dropouts, further shedding light on the relationship between the two constructs and adherence/attrition. Furthermore, questions remain concerning the cause of the higher enjoyment levels found among those who used a variable training method. Potential reasons for why enjoyment increased under these conditions and whether enjoyment was a direct cause of increasing adherence should be explored, and additional conditions that involve regular exercisers may reveal interesting findings in this regard.

It is reasonable to assume that alternative variable schedules could differentially influence adherence rates. Perhaps an ideal variable schedule would be one in which the exercise mode is varied daily, weekly, or monthly, as opposed to bimonthly. Investigation of other variable schedules is certainly warranted. That is, the transition between exercises is an area for future research for a number of reasons beyond adherence effectiveness. For example, inappropriate shifts in exercise mode could increase the risk of injury. Finally, more sensitive fitness measures, such as VO2max could be utilized to gain a more accurate portrayal of fitness levels both before and after the training period. Systematic examination of these issues may assist fitness specialists to develop new guidelines to help people adhere to their exercise programs.

The implications of this initial attempt to understand variability in exercise are potentially significant. Revealing that varying a workout leads to greater frequency of exercise can be important, whether it is for personal or medical reasons, to those who have difficulty adhering to a fitness program. If further substantiation for implementing variable exercise programs is provided, guidelines can be recommended to the general population to assist in achieving fitness goals. This mission is in line with the broader goal of assisting people to exercise more frequently, thereby improving their health and well-being.

Authors' Note

The authors would like to thank Heather Hausenblas, Jim Cauraugh, William Vitulli and an anonymous reviewer for their insightful comments on an early version of this manuscript. In addition, we would like to express our sincere appreciation to Kevin Taylor for his help with the statistical analyses for this project.


Bandura, A. (1997). Self-efficacy: The exercise of control. New York, NY: W.H. Freeman & Co.

Blair, S.N,, Kohl, H.W., Paffenbarger, R.S., Clark, D.G., & Cooper, K.H. (1989). Physical fitness and all-cause mortality: A prospective study of healthy men and women. Journal of the American Medical Association, 262, 2395-2401.

Borg, N. (1985). An introduction to Borg's RPE scale. Ithaca, NY: Mouvement.

Boyer, J.L., & Kasch, F.W. (1970). Exercise therapy in hypertensive men. Journal of the American Medical Association, 211, 1168-1171.

Brownell, K.D., & Stunkard, A.J. (1980). Physical activity in the development and control of obesity. In A.J. Stunkard & W.B. Saunders (Eds.), Obesity (pp. 106-141). Philadelphia, PA: Saunders.

Camacho, T.C., Roberts, R.E., Lazarus, N.B., Kaplan, G.A., & Cohen, R.D. (1990). Physical activity and depression: Evidence from the Alameda County study. American Journal of Epidemiology, 134, 220-231.

Cardinal, B.J., & Sachs, M.L. (1995). Prospective analysis of stage-of-exercise movement following mail-delivered, self-instructional exercise packets. American Journal of Health Promotion, 9, 430-432.

Cardinal, B.J., & Sachs, M.L. (1996). Effects of mail-mediated, stage-matched exercise behavior change strategies on female adults' leisure-time exercise behavior. Journal of Sports Medicine and Physical Fitness, 36, 100-107.

Crocker, P., Bouffard, M., & Gessaroli, M. (1995). Measuring enjoyment in youth sport settings: A confirmatory factor analysis of the physical activity enjoyment scale. Journal of Sport & Exercise Psychology 17, 200-205.

Dishman, R.K. (Ed.). (1988). Exercise adherence: Its impact on public health. Champaign, IL: Human Kinetics.

Dishman, R.K., & Buckworth, J. (1997). Adherence to physical activity. In W.P. Morgan (Ed.), Physical activity and mental health (pp. 63-90). Washington, DC: Taylor & Francis.

Gallucci, N.T. (1995). Effects of goals on the maintenance of exercise programs. Journal of Sport Behavior, 18, 109-122.

Golding, L.A., Myers, C.R., & Sinning W.E. (Eds). (1989). Y's way to physical fitness: The complete guide to fitness testing and instruction. Champaign, IL: Human Kinetics.

Ingledew, D., Markland, D., & Medley, A. (1998). Exercise motives and stages of change. Journal of Health Psychology 3, 477-489.

Kendzierski, D., & DeCarlo, K. (1991). Physical activity enjoyment scales: Two validation studies. Journal of Sport & Exercise Psychology, 13, 50-64.

Kimiecik, J.C., & Harris, A.T. (1996). What is enjoyment? A conceptual/definitional analysis with implications for sport and exercise psychology. Journal of Sport & Exercise Psychology 18, 247-263.

King, A.C., Haskell, W.L., Taylor, C.B., Kraemer, H.C., & DeBusk, R.F. (1991). Group- vs. home-based exercise training in healthy older men and women: A community-based clinical trial. Journal of the American Medical Association, 266, 1535-1542.

King, A.C., Haskell, W.L., Young, D.R., Oka, R.K., & Stefanick, M.L. (1995). Long-term effects of varying intensities and formats of physical activity on participation rates, fitness, and lipoproteins in men and women aged 50 to 65. Circulation, 91, 2596-2604.

King, A.C., Kiernan, M., Oman, R.F., Kraemer, H.C., Hull, M., & Ahn, D. (1997). Can we identify who will adhere to long-term physical activity? Signal detection methodology as a potential aid to clinical decision making. Health Psychology 16, 380-389.

King, A.C., Taylor, C.B., Haskell, W.L., & DeBusk, R.F. (1990). Identifying strategies for increasing employee physical activity levels: Findings from the Stanford/Lockheed exercise survey. Health Education Quarterly 17, 269-285.

Lombard, D.N., Lombard, T.N., & Winett, R.A. (1995). Walking to meet health guidelines: The effect of prompting frequency and prompt structure. Health Psychology 14, 164-170.

Londeree, B.R., & Moeschberger, M.L. (1984). Influence of age and other factors on maximal heart rate. Journal of Cardiac Rehabilitation, 4,44-49.

Marcus, B.H., & Stanton, A.L. (1993). Evaluation of relapse prevention and reinforcement interventions to promote exercise adherence in sedentary females. Research Quarterly for Exercise and Sport, 64,447-452.

Martin, J.E., Dubbert, P.M., Kattell, A.D., Thompson, J.K., Raczynski, J.R., Lake, M., Smith, P.O., Webster, J.S., Sikora, T., & Cohen, R.E. (1984). Behavioral control of exercise in sedentary adults. Journal of Consulting and Clinical Psychology, 52, 795-811.

McAuley, E. (1992). The role of efficacy cognitions in the prediction of exercise behavior in middle-aged adults. Journal of Behavioral Medicine, 15, 65-88.

McAuley, E., Courneya, K.S., Rudolph, D.L., & Lox, C.L. (1994). Enhancing exercise adherence in middle-aged males and females. Preventive Medicine, 23, 498-506.

McAuley, E., & Jacobson, L. (1991). Self-efficacy and exercise participation in sedentary adult females. American Journal of Health Promotion, 5, 185-207.

McAuley, E., Lox, C., & Duncan, T.E. (1993). Long-term maintenance of exercise, self-efficacy, and physiological changes in older adults. Journal of Gerontology 48, 218-224.

Morgan, P.P., Shephard, R.J., Finucane, R., Schimmelfing, L., & Jazmaji, V. (1984). Health beliefs and exercise habits in an employee fitness programme. Canadian Journal of Applied Sport Sciences, 9, 87-93.

Myers, J.L., & Well, A.D. (1995). Research design and statistical analysis. Hillsdale, NJ: Lawrence Erlbaum.

Neale, A.V., Singleton, S.P., Dupuis, M.H., & Hess, J.W. (1990). The use of behavioral contracting to increase exercise activity. American Journal of Health Promotion, 4, 441-447.

Oberman, A. (1985). Exercise and the primary prevention of cardiovascular disease. American Journal of Cardiology, 55, 10-20.

Perkins, K.A., Rapp, S.R., Carlson, C.R., & Wallace, C.E. (1986). A behavioral intervention to increase exercise among nursing home residents. The Gerontologist, 26,479-481.

Perri, M.G., Martin, A.D., Leermakers, E.A., Sears, S.F., & Notelovitz, M. (1997). Effects of group- versus home-based exercise in the treatment of obesity. Journal of Consulting and Clinical Psychology, 65, 278-285.

Petruzzello, S.J., Landers, D.M., Hatfield, B.D., Kubitz, K.A., & Salazar, W. (1991). A meta-analysis on the anxiety-reducing effects of acute and chronic exercise. Sports Medicine, 11, 143-182.

Pollock, M.L., Gaesser, G.A., Butcher, J.D., Despres, J., Dishman, R.K., Franklin, B.A., & Garber, C.E. (1998). The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness, and flexibility in healthy adults. Medicine and Science in Sports and Exercise, 30, 975-991.

Powers, S.K, & Howley, E.T. (1997). Exercise physiology: Theory and application to fitness and performance (3rd ed.). Dubuque: Brown and Benchmark.

Robison, J.I., Rogers, M.A., Carlson, J.J., Mavis, B.E., Stachnik, T., Stoffelmayr, B., Sprague, H.A., McGrew, C.R., & VanHuss, W.D. (1992). Effects of a 6-month incentive-based exercise program on adherence and work capacity. Medicine and Science in Sport and Exercise, 24, 85-93.

Rodgers, W.M., & Gauvin, L. (1998). Heterogeneity of incentives for physical activity and self-efficacy in highly active and moderately active women exercisers. Journal of Applied Social Psychology, 28, 1016-1029.

Ryckman, R.M., Robbins, M.A., Thornton, B.,& Cantrell, P. (1982). Development and validation of a physical self-efficacy scale. Journal of Personality and Social Psychology, 42, 891-900.

Simkin, L.R., & Gross, A.M. (1994). Assessment of coping with high-risk situations for exercise relapse among healthy women. Health Psychology 13, 274-277.

Spence, J.C., Poon, P., & Dyck, P. (1997). The effect of physical-activity participation on self-concept: A meta-analysis. Journal of Sport & Exercise Psychology, 19, S 109.

U.S. Dept. of Health and Human Services, Centers for Disease Control and Prevention, ational Center for Chronic Disease Prevention and Health Promotion, President's Council on Physical Fitness and Sports. (1996). Physical activity and health: A report of the Surgeon General. Atlanta, GA: International Medical Publishing.

Vitulli, W.F., & DePace, A.N. (1992). Manifest reasons for jogging and not jogging. Perceptual and Motor Skills, 75, 111-114.

Vranic, M., & Berger, M. (1979). Exercise and diabetes. Diabetes, 28, 147-167.

Wankel, L.M. (1993). The importance of enjoyment to adherence and psychological benefits from physical activity. International Journal of Sport Psychology 24, 151-169.

Wankel, L.M. (1997). "Strawpersons," selective reporting, and inconsistent logic: A response to Kimiecik and Harris's analysis of enjoyment. Journal of Sport & Exercise Psychology, 19, 98-109.

Weber, J., & Wertheim, E.H. (1989). Relationships of self-monitoring, special attention, body fat percent, and self-motivation to attendance at a community gymnasium. Journal of Sport & Exercise Psychology, 11, 105-114.

Welsh, M.C., Labbe, E.E., & Delaney, D. (1991). Cognitive strategies and personality variables in adherence to exercise. Psychological Reports, 68, 1327-1335.
 Participant Characteristics across Groups
 Variable Static Preferred Mode
Men 5 2 6
Women 19 20 9
Age 20.67 [+ or -] 2.18 20.91 [+ or -] 4.25 22.20 [+ or -] 3.51
 Means and Standard Deviations (SD)
 for All Dependent Variables
 Variable (N=24) Static (N=22)
 Mean SD Mean SD
Total Adherence 15.00 [a] 8.65 13.00 [a,b] 8.05
Ideal Adherence 2.75 [a] 2.13 2.27 [a] 2.12
Enjoyment 2.13 [a] .97 2.56 [a,b] .833
Boredom 5.54 [a] 1.05 5.18 [a] .70
PSE Pretest 78.79 [a] 7.63 78.14 [a] 6.64
PSE Posttest 78.17 [a] 6.55 76.09 [a] 7.86
Recovery Heart Rate
 Pretest 120.71 [a] 15.62 122.23 [a] 11.51
 Posttest 112.54 [b] 13.70 113.0 [b] 13.62
Manipulation Check 3.83 [a] 1.83 3.33 [a] 2.15
RPE 14.45 [a] 1.55 14.42 [a] 1.79
 Preferred (N=15)
 Mean SD
Total Adherence 9.20 [b,c] 8.49
Ideal Adherence 1.73 [a] 2.19
Enjoyment .09 [b,c] 1.11
Boredom 5.04 [a] 1.23
PSE Pretest 75.93 [a] 9.11
PSE Posttest 76.13 [a] 9.21
Recovery Heart Rate
 Pretest 125.40 [a] 17.99
 Posttest 118.67 [b] 17.98
Manipulation Check 3.07 [a] 2.37
RPE 14.38 [a] 1.60
Note. Means having common superscripts do not differ significantly
according to planned contrasts and post-hoc analyses.
COPYRIGHT 2001 University of South Alabama
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Glaros, Nicole M.; Janelle, Christopher M.
Publication:Journal of Sport Behavior
Article Type:Statistical Data Included
Geographic Code:1USA
Date:Mar 1, 2001
Previous Article:Flexible Short-term Goals and Basketball Shooting Performance.
Next Article:Athletic Injury Experience: A Qualitative Focus Group Approach.

Related Articles
Exercise as a counseling intervention. (Theory).
Low-intensity exercise, three times per week, of little cardiac benefit. (Study of 492 Sedentary Adults).
Older adults with arthritis: exercise perceptions.
Applying the transtheoretical model to exercise adherence in clinical settings.
Influence of home exercise performance, concurrent physical activities and analgesics on pain in people with osteoarthritis.
Diabetes & lifestyle.
Inactivity, Exercise Training and Detraining, and Plasma Lipoproteins, STRRIDE: A Randomized, Controlled Study of Exercise Intensity and Amount.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters