Beliefs Differentiating Levels of Exercise Adherance in Males Age 30 and Older

Beliefs were studied in five groups of individuals representing different le~els or stages of adherence to exercise. The five groups were immotives, contemplators, recruits, adherers and dropouts. Based on Fishbein's model of behavioral intention, a survey instrument was developed. First, 23 exercise professionals were interviewed for what they had found people to believe about exercise. A 69-item belief survey was formulated from this information, and it was administered to 220 males. Additionally, a background questionnaire was administered, evaluating exercise behavior and other demographic variables. An analysis of variance was conducted, and comparisons were made between pairs of the five groups. Most significant comparisons involved the immotive group. Time and mental health benefits distinguished almost all groups, with adherers relating most negatively to the fact that time is a problem and most positively to the mental health benefits that exercise provides. Immotives were opposite in nature. Recruits and adherers were found to think quite similarly.

So, why is physical activity participation so low?
Or, better yet, why do some people exercise regularly, some inconsistently, and some people refrain from exercise altogether?
As a health educator, it is this author's concern to study the belief patterns of exercisers as compared to other groups of individuals that do not exercise enough in an effort to find out what focus professionals need to take in educating individuals about the benefits of exercise so that they will undertake and maintain the behavior.
The Fishbein model of behavioral intention suggests that beliefs are the underlying key to behavior (Azjen and Fishbein, 1980). The research of Mcconnaughy, Prochaska, and Velicer (1980) delineates five stages of change that individuals pass through in implementing a behavior change.
Referencing these researchers, this study attempted to identify differences in beliefs which may exist among the following groups: those that have no intention of exercising (immotives), those that are thinking of starting -3starting an exercise program (contemplators), those that have recently started an exercise program (recruits), and those that have been maintaining a regular exercise program for at least two years (adherers), as well as those that have dropped out of exercise programs (dropouts).
It is believed that through such research, health professionals will better understand and be better prepared to motivate the American population to exercise.

Statement of the Problem
The Fishbein model of behavioral intention (Azjen and Fishbein, 1980), which posits that actual behavior is a function of behavioral intention, has recently gained acceptance in predicting health behaviors. Some applications have been with weight loss (Saltzer, 1981), alcolhol consumption (Beck, 1981), and exercise participation (Riddle, 1980). The model sta~es that behavioral intention is determined by an attitudinal component, as well as a social component, and that each of these can be further reduced into subcomponents.
The attitudinal component can be segmented into (1) beliefs about an attitude object and (2) evaluation of the consequences of this belief. The social component can be reduced to (l) a perception of beliefs that significant others have toward the attitude object and (2) the person's motivation to comply with those referents . McConnaughy, et al. (1980) have delineated five stages associated with behavior change that can be used in com-J bination with Fishbein's model to study beliefs in different exercise groups.
The purpose of this study was threefold. First, the research was undertaken to explore the possible beliefs that could affect the behavioral intention of exercising through their influence of attitudes. Second, it was undertaken to expand the purview of previous studies to include various types of exercise. And third, it was undertaken to compare the beliefs concerning exercise with five groups. These groups were immotives, contemplators, recruits, adherers, and dropouts. The purpose was considered to be one of identifyin-g differences in specific beliefs between these groups. Studies examining the reliability of beliefs of these groups is recommended for future investigation.
The author first needed to gather information from the general public regarding their beliefs about exercise. This could be misconception or fact. This was in order to develop a pool of items to be evaluated. It was felt, however, that if the public was interviewed directly, interviewees might not be sincere in their responses.
If they were only to relate concepts that they felt were universally accepted, rather than those representative of their true ideas, a wealth of material would be overlooked. It was, therefore, concluded that if the professionals that spent time with the public in exercise situations were interviewed in regard to what they had heard people say they believe about exercise, a more accurate account of the public beliefs would be generated.
After interviewing exercise professionals for the information, the survey instrument was developed according to specifications (Azjen and Fishbein, 1980). Along with a background questionnaire that measured exercise participation history and other demographic information the survey was admininistered to 267 males over the age of 30.
Subjects were grouped into one of the five exercise categories dependent upon behavior reported in the background questionnaire, and items were then analyzed for similarities and differences between pairs of groups.
Justification for and Significance . of the Study The objective of health educators is : to influence health behavior (Rash & Pigg, 1979). To be considered successful, however, such an influence must result in permanent changes -maintenance of the health behavior.
Substantiating the need for an emphasis on exercise behavior particularly, Pollock (Pollock, Wilmore, and Fox, 1979) states that physical activity may reduce the occurrence or severity of coronary heart disease by in-£reasing collateral vascularization, vessel size, myocardial efficiency, efficiency of peripheral blood distribution and return, tolerance to stress, and red blood cell mass and volume, and by decreasing serum lipid levels, obesity, arterial blood pressure and heart rate.
Consistent with such research, Froelicher & Oberman (1972) have discussed the risks associated with sedentary living. Further, exercise adherence seems to .be presenting a challenge. Dishman (1980) has found that roughly half of initial participants discontinue programs within the first six months of involvement. Even more alarming, Kentala (1972) reports the dropout statistic as high as 87% in a cardiac rehabilitation program for post infarction patients.
Exercise-adherence research has basically followed two paths. One has been to analyze the effects of situa- In relation to exercise, Riddle (1980) andSonstroem (1981) have already examined the efficacy of the model.  (Dishman, 1980) .
In part, it has been found that the reason for such poor adherence can be explained through situational factors such as the distance of the individual from the exercise facility or how the spouse thinks or feels about exercise (Andrew, et al., 1981). But, additionally, research on more personal factors, such as attitudes needs to be carried out.

Early Attitudinal Research
Attitudes are, basically, underlying variables that guide or influence behavior. As such, they are subjective in nature and offer difficulty in measurement.
Some early measurement techniques in the field of attitudinal theory made use of various scales . One such scale was developed by L.L. Thurstone. According to Thurstone, attitudes could be measured in the following manner .
First, the attitude was specified. A collection of a wide variety of opinions relating to the attitude was then carried out. The information was edited to form approximately 100 statements of opinion . The statements were scaled by a few hundred people . Some were disregarded due to ambivalence or i rrel evence ·, and an atti tutde seal e resulted from the remaining statements. Attitudes were then measured by requesting that participants either endorse (indicated by a + stgn) or reject (indicated by asign) each statement. The score was the average value of all the statements he/she had endorsed . As is o~vious, the formulation of such a scale was quite time consuming (Fishbein, 1967 (Fishbein, 1967).
According to Azjen and Fishbein (1980), it was Allport, however, who finally pointed out that attitudes were made up of both a cognitive, as well as an affective component, and it was Charles Osgood who recognized that the semantic differential scale, one that scales . from one extreme to the other, offered mor~ sensitivity in measurement than the two-point system .
In attempting to measure health-related attitudes as they relate to behavior, Jaccard (1975) pointed out two major problems -the first being the identification of those variables that determine behavioral intention in the specific area of health, and the second being the extent to which the intentions are predictive of behavior . Symbolically, the model is represented as follows: Where B is the behavior, BI is behavioral intention, · attitude toward the behavior, SN is the subjective As1s norm, and w 1 and w 2 are the empirically determined weights (Azjen and Fishbein, 1980 In measuring attitude, thus, the sum of all beliefs multiplied by their evaluative worth is symbolically represented as follows: where bi is beliefs 11 i 11 about the o~ject, ei is the evaluative aspect of bp and N is the number of beliefs (Azjen and Fishbein,l980).
The belief items are measured on a semantic differential scale which is bipolar in nature as shown below: My participation in a regular program of exercise would make me lose weight .
LI KE LY UNLIKELY extremely quite slightly neither slightly quite extremely Subjects respond as to whether they believe the statement is likely or unlikely. Similarly, evaluation items are measured on the same type of scale as shown below: My losing weight would be . . .

GOOD BAD
extremely quite slightly neither slightly quite extremefy Subjects respond as to whether they believe the consequence is good or bad.
To insure accurate measurement of the behavioral intention of concern, Fishbein emphasizes the importance of specification of four behavioral elements in each statement. These are action, either a single behavior or a behavioral -category; target, what the behavior is directed toward; context, in what situations the behavior will take place; and time, when the behavior will be carried out.
To illustrate this, the following example is provided.
If we were measuring an individua_ l 's beliefs regarEling drinking, the action would be the drinking behavior, the target would be thw substance drunk, the context would be the occasions of drinking, and the time would be the period measured. Thus, a statement to be used to measure the beliefs regarding the behavioral intention of concern would begin as follows: My drinking alcohbli c beverages in any location within the next two weeks. (TIME) The absence of any of the elements in the statement reduces the predictive accuracy of that statement. If the target were omitted, for example, we would be measuring all drinking behavior. That would include soda, milk, water, etc. If we omitted the context, we would be unsure of the scope of our behavioral measurement. This would also be true with the omission of the time element .
Ap~lications of the Model to Health Education

Weight Control
Applications of the Fishbein model to health education research have been made. In her study on weight loss, Saltzer (1981) concluded that behavioral intention predicted behavior for internally-oriented subjects, but for those with high measures of external control, behavioral intentions were unrelated to behavior . The study combined the research on behavioral intention with that of locus of control.
In another weight-loss study, Sejwacz, Azjen, and Fishbein (1980) evaluated 88 women over a two-month period to examine the relationship of intention to behavior.
At the outset, participants were questioned as to their attitudes toward dieting and exercise, which were determined to be the two behaviors correlated with weight loss. The behavioral measurement consisted of five specific dieting techniques representative of dieting behavior in general and three indices of physical activity representative of exercising in general.
For dieting, the techniques were: l. Avoid snacking between meals and in the evening.
2. Cut down on all starchy foods.
3. Avoid being in places where one might be tempted to eat starchy food and/or eat too much.
4. Decrease food intake in general by eating lighter meals, not having seconds, and not overeating.

Eat on a consistent and regular schedule.
For exercise, the indices were: l. Avoid long periods of physical inactivity.

Do exercises on a regula-r basis.
3. Participate in sports on a regular basis.
These behaviors were measured weekly as was weight The direct measure of intention to lose weight correlated with the intentional indices just previously mentioned (r = .60 for dieting and r = .66 for physical -18activity). Both the general measures of intention and the intentional indices permitted prediction of the corresponding behavior, though the index of dieting intention (five behaviors) was more predictive of dieting behavior (r = .60) than the direct measure of intention (r = .40).
The same was true of the index of intention to perform physical activity (r = .60) as compared to the general intention (r = .45). Performance was found to be related to intention.

Family Planning
In a study ori family planning (Azjen and Fishbein, 1980), a relationship was found between intention to have a child and behavior. As expected, women that evaluated be 1 i e f s such as 11 1 ea di n g to 1 a ck of time 11 and 11 inter fer i n g with plans" negatively were less likely to have a child.
Further, women were more likely to have a child if they (1) felt they could afford a child, (2) felt they were at a good age for child bearing, and (3) felt that the addition would make the family stronger.
In a similar study on the use of the pill, attitudes were found to be highly predictive of behavior. Questions fell into one of three general categories -physical effects, morality, and effectiveness. The more confident the tested women were of the low-percentage chance of negative outcomes associated with the use of the pill, the more likely they were to be found using it. Advantage-disadvantage beliefs were quite important. Interestingly, it was found that college women were more influenced by the attitudinal aspect of intention, whereas married women were more influenced by the normative component (Azjen and Fishbein, 1980).

Alcohol
In a study on driving under the influence of alcohol, These are (1) seriousness of the consequences associated with a behavior, (2) susceptibility to the consequences provided no action is taken to avoid them, and (3) the effectiveness of a course of action at being able to avoid these consequences. In the study, attitude was more predictive of behavior than subjective norm, and intention was more predictive of behavior than attitude . The Health Belief Model was less successful in predicting intention to drive while under the influence of alcohol (Beck, 1981).

Exercise
Recently, the Fishbein model was applied to two studies of the prediction of exercise participation . Sonstroem (1) immotives -non-exercisers who have no intention of taking part in the behavior. If they are involved in a program, it is probably out of force from a spouse or family member; (2) contemplatorsthose that are considering participation in the behavior.
They are struggling to understand the problem. They want more information, but are not yet committed to change; (3) recruits -those that have recently started taking part in the behavior . They feel ready for change, willing to invest to make change, but have not yet reached a maintenance of change of the behavior; and (4) adherersthose that regularly participate in the behavior but may still be struggling with the possibility of relapse.
In addition, a fifth group, dropouts, appear to complete the cycle that Prochaska and DiClimente (1981) have documented to exist in behavior change therapy .
According to Prochaska, individuals tend to move through stages one to four in a cyclical fashion relapsing occasionally and starting.back at the beginning. They then move through the same stages again. Additionally, Prochaska has found participants to drop out of therapy at all stages, thus breaking away from the cycle. The resultant dropout group consists of participants at all levels of behavior change. These same participants frequently return repeatedly, moving through the cycle in the same pattern. Mcconnaughy, et al. (1980) developed a 32-item inventory which proved highly predictive in delineating stages of change in smoking cessation and maintenance .
As they point out, by evaluating the stage of change a client is in, readiness for change can be ascertained, and the appropriate processes of change can be matched to the level that the client is at, producing a more systematic approach to psychotherapy.
For example, if a subject is in stage 3, recruit, he may be very turned off by talk but much more receptive to behavior change techniques. Immotives, however, may be quite far from behavior change readiness but quite receptive to information.
Since exercise participation has been shown to assume essentially the same attrition curve over time as medical and psychological therapies (Dishman, et al . , 1980), it is assumed that the psychological processes involved may be similar, thus applicable to the present study. Collection is reviewed. This consisted of the pilot study, the sample, and the method of administration.
Finally, the statistical Analyses are discussed.

Inventory Development
Belief Questionnaire The belief questio~naire consisted of 69 items developed from information gained as described below.
Development of Item Pool. The first stage of the research consisted of tbe development of the items for the attitude survey. This process began with interviews of professionals in the field of exercise. The rationale for this is described in Chapter l.   extremely quite slightly neither slightly quite extremely A score of 0 supported the statement most highly (extremely likely -extremely good), and a score of six rejected the ;tern completely (extremely unlikely -extremely bad) .
Illustrating the scoring systme, a subject that responded "slightly unlikely" to a belief statemnt received a score of 11 4 11 for that statement.
Background Questionnaire The background questionnaire was utilized to differentiate groups to be analyzed. According to responses on each of the eight questions, subjects were grouped in regard to their exercise behavior, their socioeconomic status, their weight, their smoking habits and their sex.
Method of Scoring. According to responses to questions one, two, and three, participants were categorized into one of five groups regarding their exercise behaviors.
The five groups were immotives, contemplators, recruits, adhereres, and dropouts .  Though some of the community and rotary groups did comply, their total participation accounted for less than a quarter of the data.

Test Instructions
--Participants were instructed to read and sign the informed consent sheet, read directions carefully, and complete~ survey items. They were told that this information was being gathered so that the author could learn more about people's beliefs concerning exercise.
They were informed that there were !!_Q_ right or wrong answers and that they should respond according to the way they truly feel in regard to each survey item. Individuals with questions regarding the completion of the instrument were referred back to the direction sheet.
No time limit was given for completion of the survey.

Data Deletion
Tests were disregarded if any of the following conditions existed: Lack of variation -if one full page showed no variation in response.
Lack of independence -if three items or less were different on the entire test.
Missing data -if more than five items were incomplete.
Medical clearance -if a subject was refraining from exercise for medical reasons.
Of the 267 surveys collected, 47 were disregarded for one of the above reasons, leaving 220 for evaluation. to outline the items that collectively predicted most accurately exercise groupings. -32-

CHAPTER IV -RESULTS
The Results section is divided into six major areas.
These areas and their sub-sections are outlined below.
Outline of the Chapter The first section involves the group comparisons.
The next section provided the results of the dis-

Analysis of Variance
Group Comparisons Table 4-1 presents a summary of the belief items that differentiated any two groups of the five studied .
Any item that appears on the chart was found to differentiate with a .05 level of significance.
Subjects were classified into five groups via the system explained in Chapter 3. Following, is a description of the comparisons between groups that were made. Five comparisons were carried out.
Three of the five items that differentiated the groups were concerned with time required to exercise and whether or not it was possible to afford such time (25,48,54). In all three cases contemplators agreed more strongly than recruits that time would be a problem. Additionally, contemplators related less to the possibility of exercise giving them time by themselves to think (29).
"Time 11 was obviously the major factor at this bridge of the adherence cycle as was the focus on negative thinking.
These were, in fact, the only variations.  4. l 0 l. l 0 3. l 2 l . 04 The fourth comparison that was done was of adherers and dropouts. It was carried out to provide information regarding the quitters as compared to the maintainers.  The final groups that were compared were immotives and the combination of recruits and adherers. This was done to provide more information about the differences between the opposing ends of the behavioral spectrum.  the responses. Certain belief concept group~ were more popular than others. Certain types of ite ins were more popular with one group than others. Below is a discussion of the similarities and differences.
Item characteristics. It is important to recognize that the higher adherence groups associated more highly with positive results of exercise and less so with negative outcomes. The lower groups tended to relate less strongly to positive outcomes and more strongly with negative outcomes.
Hypothesized item tlust~rs. Referring back to Table   3-2 all items were grouped according to subject into one of 13 belief-concept groups. Of those groups, the areas of Pain-discomfort, Prerequisites, Cost and Recreation were not found to distinguish any groups. Additionally, the areas of Eating-weight, Socializing, and Age were found to differentiate one set of groups only, immotives as compared to recruits/adherers (non-exercisers vs. exercisers). This would indicate that these six subject areas do not require emphasis in instruction but, are understood by most people better than other subject areas.
As Table 4-7 shows, the following belief-concept groups differentiated more than two groups: Disciplinecompetition, Time-commitment, Appearance, Physical-Endurance, Mental-relaxation, and self-image. It should be noted that the nine items from the Discriminant Function Analysis (DFA) included items from all six of these areas. Additionally, the Discipline-competition group was found to distinguish four of the sets of groups and the Mentalrelaxation group to distinguish all sets of groups.
With the large size of the immotive/contemplator comparison, it was possible to further analyze the beliefconcept groups that existed. In the Time-Commitment area both groups agreed with time as a problem, whereas groups 2 and 3 (contemplators, recruits) disagreed. This would indicate that recruits are the first group in the adherence cycle to lose fear of problems with time.
Two Appearance items (10,41) and five Self-image items (20,34,47,59,64) were found to differentiate these same groups. In each case, contemplators agreed more strongly With the statement than immotives. This would indicate Interestingly, during the interview phase of the study, almost each interviewee listed cosmetics as the major factor in exercising for most individuals.
The area of Physical-endurance beliefs was strong in comparing these two groups as was the Mental-relaxation area. Ten (15,19,24,30,32,37,44,46,56,62)   It should be noted that in each case, dropouts represented a strong percentage of the misclassifications.
This would suggest a pattern of some sort. This is further analyzed in the Discussion section.  -48-

Discussion
As described in Chapter 3, Prochaska and DiClemente have found behavior change to follow a sequential pattern.
Individuals tend to move through stages on a step-by-step basis from immotive to contemplator to recruit to adherer and, possibly during any stage, drop out. Accordingly, it seemed appropriate to study the differences in beliefs between sequentially-located groups.
It must be noted that much more information was found to differentiate immotives from each other group than was found to differentiate most pairs of groups, whereas the immotive comparisons consisted of approximately 20 items. This confirmed the fact that individuals at this stage are full of misconceptions regarding both costs and benefits of regular exercise.
Contemplators and recruits were differentiated in an interesting way. Four out of the ·five items differentiating the two were negatively oriented. It would seem that contemplators have already recognized benefits and are presently leery of costs of exercising.
As for the recruit/adherer relationship, the beliefs were quite similar suggesting more of an educational need at the immotive, contemplator, and dropout stages.
With the variation of the adherer/dropout table, it would seem that the dropout group was quite diversified as Prochaska and DiClimente would predict.
The overwhelming number of beliefs that differentiated 1 ;rnrnotives from recruits/adherers confirmed the theory that those that do not exercise are full of misconceptions.
Referring back to Chapter 3, page 3, Fishbein's model is described as consisting of two major componentsan attitudinal and a normative component. In this particular study only one half of one of the components (attitudinal) was evaluated. A 50.73% Prediction level was reached using only one-fourth of the model in the discriminant function analysis study.
As for the misclassifications, the fact that dropouts represented much of the error supports Prochaska and Di-Climente' s research which states that dropouts can exit from any level of the behavioral cycle. It is possible, therefore, that the misclassifications actually represent tendencies within the subjects to be ready to drop out.
Also, these misclassifications reflect similarities within groups or levels of variation . in beliefs with groups.
Comparing the results of the discriminant function analysis with that of the Analysis of Variance, several similarities were found. On Tables 4-2  The findings in this study were in one way consistent with that of both Riddle (1980) andSonstrem (1981).
In Riddle's research, as well as in the present study, exercisers (recruits and adherents) were found to relate more highly to the positive consequences and less so to the negative outcomes. The non-exercisers (immotives and contemplators) associated more strongly with negative beliefs and less so with positive outcomes. Further, both studies found exercisers to relate more highly to the mental health benefits and non-exercisers to relate more strongly to the benefits associated with time required to exercise.
In Sonstroem's study, this was also found when comparing high program attenders with low program attenders. Table 4-11 lists the similarities in results of Riddle's Sonstroem's and the present research in these two areas.
In the present, study, immotives represent non-exercisers and recruits and aherers represent exercisers for the purpose of clear comparison. In Sonstroemls study low-attenders a re c om p a re d a s n o n -e x e r c i s e r s a n d h tg;h-a t t e n d e r s a s e x e rcisers.
The major variation between the present study and the Riddle study was in the results of the Discri~ina~t Function Analysis. None of the items that accurately predicted 50.83% of the groups were found to differentiate in Riddle's study. In fact, they did not even appear in her study. It was concluded that the greater pool of items in the present study was responsible. Additional analyses with the background questionnaire revealed that adherers were significantly lighter than immotives and contemplators~ This would suggest that the exercise does result in reduced weight or that lighter people tend to adhere better to exercise, supporting the importance of prevention .
Adherers and recruits were found to smoke significantly less than immotives. This suggests an orientation to a healthy lifestyle with exercise .
Adherers, as compared to dropouts, had significantly less of a perceived medical history which keeps them from exercising. This might account for much of the dropout group.
And finally, oddly enough, immotives were found to have a significantly higher socio-economic status than three other groups, recruits, adherers, and dropouts.

PROBLEMS
Two major problems were encountered in the administration of the survey. First it was difficult finding the numbers that were required for the study. Many people refused to complete this lengthy instrument.
The second problem occurred with the instrument itself. Many individuals had a problem understanding the complex directions. Additionally, the directions were very long. Many people tried to ignore them and take the test.
Consistently, however, they turned to testing personnel for assistance. As stated earlier, 47 of the 267 completed inventories had to be discarded because of incomplete or faulty responses.  You are being asked to answer some questions on the following pages. The purpose of these questions is to develop some ideas about what people think about physical activity and what they think about themselves in relation to physical activity. This type of information will be helpful in planning exercise programs for people.
It should be emphasized that there are n-0 right or wrong, ~good or bad responses to the questions you are answering. Please answer each item as YQJ! really feel about it.
Data will remain strictly confidential. All data will be coded and will not be identified by name. You do not have to complete these inventories and you may refuse to do so at any time. Please feel free to ask any questions you may have.

Date
Name Telephone Robert J. Sonstroem

Voluntary Consent
The purpose of my participation in these procedures has been explained to me. I freely consent to participate. I understand that I am not required to participate and that I may stop participation at any time. Please describe precisely the type of work you do, your position, and the size of the company for which you work.
-60-Please indicate with a checkmark the highest level of formal schooling you received as described below: Did not complete the seventh grade.
Completed tenth or eleventh grade, but not twelve.
Completed high school.
Completed at least one year of college, but not a full college course.
Completed a four-year college or university course leading to a recognized college degree.
Completed a recognized professional course leading to a graduate degree. the questionnaire you are about to complete we ask questions ;ch make use of rating scales with seven places; you are to make check mark in the place that best describes your own opinion. r example, if you believed that the next statement was generally ue, you would respond as follows: The weather in Rhode Island is cold in January. xr, in our inventory we wi 11 take a portion of the sentence you ve just responded to and ask you to indicate whether the idea is ood 11 or 11 bad 11 • As an example: cold weather in January GOOD : : : X BAD extremely quite-slightly neither slightly quite extremely u would check the category shown above if you feel that cold ather in January is generally quite bad .
. the other hand, some people, such as skiing enthusiast s , who Joy cold weather very much, might mark the scale as follows: GOOD : : : : : BAD extremely quite slightly neither slightly quite extremely rder to reduce space, the response 11 extremely, quite, slightly, 0 neither 11 have been abbreviated to 11 _e_,_g__,_s_, and _n_" as d n below. In making your ratings, please remember the following: In this particular questionnaire we are interested in . people's views regarding participating in a regular exercise program. It should be a program that includes vigorous endurance and aerobic exercises for at least 20-30 minutes three times per week~ This could be a program conducted at home, at a healt~ spa, YMCA , or in our adult exercise progr~m at URI .
In completing this questionnaire it is not important whether you presently exercise or not . , We wantYou to think of exercise in relation t6 y6ur~~lf . Please respond to -all of the items below from the viewpoint of your own pds~ible future participation in an exercise pfogram.
For each numbered item below you are asked to indicate: A. To what extent i s the item TRUE (LIKELY) or UNTRUE (UNLIKELY) in representing your own belief, AND ....

B.
To what extent i s the idea itself either GOOD or BAD, in your opinion .          If so, please indicate the nature of this condition below.