Abstract

Relationships among full constructs of the transtheoretical model using a sample of 121 adults with mild intellectual disabilities in Taiwan were examined. Self-reports of stages of change and transtheoretical model psychosocial measures were gathered through interviews. Although MANCOVA revealed that behavioral processes of change, cognitive processes of change, self-efficacy, and perceived pros increased across stages, we did not find a clear linear pattern of association. Direct discriminant function analysis indicated that the most important predictors of stages of change were behavioral processes, cognitive processes, and self-efficacy. The overall stage of change classification accuracy using transtheoretical model psychosocial constructs was 56.2%. Psychosocial measures specifically developed for this population should be further explored.

Participating in moderate intensity physical activity on a regular basis provides a wide range of physical, social, and psychological health benefits for all, including those with intellectual disabilities (Taylor, Sallis, & Needle, 1985; U.S. Department of Health and Human Services, 1996, 2002; World Health Organization, 2005). Nevertheless, only 17.5% to 33% of adults with intellectual disabilities meet physical activity guidelines (Temple, Frey, & Stanish, 2006). Increasing physical activity in this underserved and understudied population is considered a key priority in public health and education (U.S. Department of Health and Human Services, 2005).

Because most investigations on physical activity adoption and adherence have been nontheoretical, researchers have been urged to use validated theory-driven interventions when examining physical activity behavior (e.g., Martin & Dubbert, 1982; Sallis & Hovell, 1990). The transtheoretical model of behavior change has received extensive attention with regard to adoption and adherence to an active life-style.

The transtheoretical model integrates four interrelated dimensions; (a) stages of change (the core construct of the model, which identifies individuals in one of the five stages of change: precontemplation, contemplation, preparation, action, and maintenance); (b) processes of change (including cognitive and behavioral strategies that individuals can use to promote change in different stages); (c) decisional balance (individuals' weighting of the perceived pros and perceived cons in different stages); and (d) self-efficacy (individuals' situational confidence in performing and maintaining the different stages of change) (Prochaska & Velicer, 1997a). Thus, the stages of change represent when an individual is ready to change, whereas the processes of change illustrate how an individual will make changes. Self-efficacy refers to the level of situational confidence that individuals feel when making the changes. Finally, decisional balance considers the desired or undesired consequences that result from changing a behavior (Prochaska & Velicer, 1997b). In order for the intervention to be effective, transtheoretical model proponents advocate that adequate interventions need to be introduced to an individual at the right time and with the right approach. The model leads developers of physical activity programs to use a stage-matched intervention instead of one universal intervention. For example, for people who have no intention of becoming regularly physically active, health education classes may be more effective than prematurely moving them into an exercise program.

The applicability of the transtheoretical model to promote regular physical activity has been documented across many different populations without disability and the population with physical disabilities (e.g., Cardinal, Kosma, & McCubbin, 2004; Dunn, 1996; Marcus, King, Clark, Pinto, & Bock, 1996; Marshall & Biddle, 2001). Its applicability to the population with intellectual disabilities, however, has not been examined. Traditional physical activity interventions for this population have been focused on action-oriented exercise programs. There is little research on the application of educational and behavioral approaches to foster adoption and maintenance of regular physical activity for persons with intellectual disabilities (Heller, Hsieh, & Rimmer, 2004). Therefore, describing characteristics of such individuals in each stage of readiness and identifying factors that explain and predict their physical activity behaviors have substantial implications for developers of physical activity programs to successfully promote an active lifestyle and enhance health status for this population.

Our purpose in conducting this study was to explore the applicability of the full transtheoretical model to adults with mild intellectual disabilities. To accomplish this goal, we identified three objectives. First, we explored the relationships among stages of change and the transtheoretical model psychosocial constructs (processes of change, self-efficacy, and decisional balance). Our objective was to determine whether a theorized nearly linear pattern of association would be found between stages of change and the transtheoretical model psychosocial constructs (i.e., individuals classified in higher stages of change would exhibit higher scores in cognitive processes, behavioral processes, perceived pros, and self-efficacy as well as a lower score in perceived cons). Second, we determined classification accuracy of stages of change using the transtheoretical model psychosocial constructs. Finally, we identified the most influential transtheoretical model psychosocial factors that adequately predicted stages of change.

Method

Participants

As can be seen in Table 1, among the 121 participants with mild intellectual disabilities, the majority was young males, lived with their families, and had obtained a high school diploma. With regard to weight, 51.4% were either overweight (World Health Organization, 2000; 23 < BMI < 25) or obese (BMI > 25) and about 8% were underweight. Only a small proportion of participants had Down syndrome. (See Table 1 for detailed participants' demographics).

Table 1

Demographics of Participants

Demographics of Participants
Demographics of Participants

Measures

Screening Tests

Orientation Test. We used the Orientation Test, a 10-item scale extracted from a standard mental status examination (Folstein, Folstein, & McHugh, 1975) to assess orientation to time and place. The scores range from 0 to 10, with higher scores representing better performance.

Modified Fuld Object-Memory Test. The Modified Fuld Object-Memory Test (Seltzer, 1997) assesses recognition/naming and learning/recall. Objects (rather than pictures) are used to improve salience, and multiple presentations are included (generating a stable learning score) as is a recognition trial. We entered these two screening tests as covariates in the main data analyses to control the potential impact of participants' memory and time and place concepts on their responses (Sano, Aisen, Dalton, Andrews, & Tsai, 2005).

Stages of Change

We modified the Stage of Change Scale (Marcus & Simkin, 1993), a 5-choice response pool, for use in the classification of participants into one of the five stages of change: precontemplation, contemplation, preparation, action, and maintenance. Several modifications and adaptations (e.g., breaking down questions into a sequence of connected questions and use of photos to facilitate understanding of questions) were provided to ensure a careful response from participants with mild intellectual disabilities (the Modified Stages of Change Scale and instructions for interviews are available from the first author.) We defined regular physical activity as performing physical activities during leisure time at least 5 days per week for at least 30 min per day at a moderate or higher intensity based on the U.S. Surgeon General and American College of Sports Medicine guidelines (American College of Sports Medicine, 1990; Pate et al., 1995; U.S. Department of Health and Human Services, 1996).

Processes of Change

The 30-item Processes of Change Scale (Nigg, Norman, Rossi, & Benisovich, 1999) contains two higher order dimensions (Behavioral and Cognitive Processes), with a total of 10 factors. Cognitive Processes include consciousness-raising, dramatic relief, environmental reevaluation, self-reevaluation, and social liberation. Behavioral Processes include counter-conditioning, helping relationships, reinforcement management, self-liberation, and stimulus control. To accommodate participants with mild intellectual disabilities, we adapted the 5-point Likert scale to a 4-point pictorial Likert-type scale (0  =  not applicable/don't know; 1  =  never, 2  =  sometimes, 3  =  always) (Finlay & Lyons, 2001).

Decisional Balance

We used the 10-item Decisional Balance Scale (Plotnikoff, Blanchard, Hotz, & Rhodes, 2001), which includes two subscales, Perceived Pros and Cons. Each of these subscales contains five items. A sample Perceived Pro item is “Physical activity would help me control my weight,” and a sample Perceived Con item is “I'd worry about looking awkward if others saw me being physically active.” This scale was also modified from the original 5-point Likert scale into a 4-point pictorial Likert type scale (0  =  not applicable/don't know, 1 =  not at all, 2  =  fairly, 3  =  very much).

Self-Efficacy

The Self-Efficacy Scale (Benisovich, Rossi, Norman, & Nigg, 1998), an 18-item self-report inventory, is used to assess one's situational confidence in performing and maintaining the change. The scale contains 6 subscales: (a) Negative Effects, (b) Excuse-Making, (c) Exercising Alone, (d) Access to Equipment, (e) Resistance From Others, and (f) Weather. A sample item on this scale assessing excuse-making is “I'm confident that I can participate in physical activity when I am busy.” Participants respond to each item using a 4-point Likert type scale (0  =  not applicable/don't know, 1  =  not confident, 2  =  not sure, 3  =  confident).

Procedure

Pilot Study

We used the cross-cultural translation technique to translate all instruments in the study (Banville & Desrosiers, 2000; Vallerand, 1989). Six primary caregivers, who had communicated closely with persons who have intellectual disabilities, reviewed and revised the wording of all psychosocial items. Two content experts reviewed all the translations to determine the content validity and readability of the Chinese instruments for Taiwanese adults with mild intellectual disabilities. Both content experts were university faculty members, had content expertise in the transtheoretical model and physical activity, and were conversant in both Chinese and English. After the translations were approved by the two content experts, we recruited 20 bilingual college students who were proficient in both Chinese and English. Within a 2-week interval, they responded to both the Chinese and English questionnaires in order to evaluate the concurrent validity of the Chinese questionnaires. We translated and implemented a brief screening test developed by Vallerand and Halliwell (1983) to determine their eligibility. The concurrent and content validity and reliability of the translated instruments were confirmed before a pilot study was conducted.

In the pilot study, we implemented all of the measures on 10 participants with mild intellectual disabilities who were selected intentionally to include those with mild intellectual disabilities at the lowest threshold (i.e., borderline mild–moderate). Internal consistency of the 10 participants' responses on the transtheoretical model psychosocial measures were examined using Cronbach's alpha (Processes of Change, α  =  .940; Self-Efficacy, α  =  .917; Decisional Balance, α < .688). Feasibility of implementation and time estimations of all the measures, as well interinterviewer reliability, were evaluated (see Interviewer Training and Interinterviewer Reliability section below). We made necessary modifications after the pilot study.

Interviewer Training and Interinterviewer Reliability

The first author and 5 research assistants served as interviewers. Interview training sessions were conducted before the pilot study and included the following sections: (a) an overview of the research procedures, (b) a review of interview instruments, (c) a review of the interview techniques for adults with mild intellectual disabilities, and (d) interview simulations. The first author maintained close communication with the research assistants throughout the study. We examined interinterviewer reliability during the pilot study. The 6 interviewers were numbered and paired for the interinterviewer reliability test. Each of the 10 participants was interviewed separately by 2 interviewers, with a 10-min interval between interviewers. Average intra-class correlation coefficients (ICC) between 2 interviewers ranged from .618 to .989 (Behavioral Processes of Change, ICC  =  .989; Cognitive Processes of Change, ICC  =  .913; Decisional Balance, ICC  =  .735; Self-Efficacy, ICC  =  .926). Overall, the interinterviewer reliability was acceptable.

Participant Recruitment and Selection Procedures

We initially recruited 185 adults with mild intellectual disabilities from the headquarters of the Parents' Association for Persons With Intellectual Disabilities, Special Olympics, vocational training centers, and other related organizations and institutions in four cities or counties selected from northern, central, southern, and eastern Taiwan. Participant eligibility criteria included Taiwanese adults who (a) were ambulatory, (b) were at least 18 years of age, (c) received an official identification of intellectual disability, (d) possessed the receptive and expressive communication skills necessary to follow simple directions, (e) produced understandable responses, and (f) could recall events from the past 2 weeks. These determinations were made by the staff members of the service agencies contacted or by the primary caregivers. Written informed consents for participants with mild intellectual disabilities and their primary caregivers or guardians were obtained before beginning the interviews. Out of the 185 potential participants, 156 attended an interview and 29 were absent due to personal issues (e.g., sickness, changing their mind about participation) and/or bad weather conditions, resulting in a participation rate of 84.3%.

We screened participants for the following three exclusions. First, before conducting formal survey interviews, we adopted and implemented two screening tests: (a) Orientation to Time and Place (Folstein et al., 1975) and (b) the Modified Fuld Object-Memory Tests (Seltzer, 1997). We employed these tests to screen for participants' concept in time and place as well as memory, which are important abilities needed to respond to interview questions. The 11 participants who were unable to complete either one of the tests were excluded from the study. In total, 145 participants completed both screening tests.

Second, trained interviewers used a 3-point rating scale (1  =  no or limited evidence of clear understanding of questions, 2  =  not sure, 3  =  obvious evidence of clear understanding of questions) to rate participants' understanding of each of the four transtheoretical model psychosocial measures. If a participant received 1 point in an interviewer rating on any measure, his or her responses on that particular measure were voided. The interviewers identified 10 participants who received 1 point in Decisional Balance, Self-Efficacy, and Processes of Change scales and 5 participants who received 1 point in the Stages of Change Scale.

Finally, we integrated three reliability questions into each of the measures (decisional balance, self-efficacy, and processes of change). These reliability questions were written using reverse wording to examine the consistency of participants' responses (Finlay & Lyons, 2001). Participants received 1 point when their response to a reliability question was consistent with its corresponding question. The full point range was from 0 (minimum) to 3 (maximum) for each of the three psychosocial measures. If a participant received a point totaling 1 or 0 in any one of three measures, his or her responses in that particular measure were excluded. The number of participants who received 1 point or less for decisional balance, self-efficacy, and processes of change were 2, 1, and 4, respectively. Based on the above exclusionary procedures, a total of 121 eligible participants who had completed responses for all the scales were included in this study.

Data Collection

We obtained approval for the protection of research participants from the Institutional Review Board prior to the data collection. The formal interviews took place in private rooms provided by the recruiting service agencies. Prior to the interviews, the interviewer went over the informed consent, using clear language that was devoid of confusing terminology. The interview began only after these adults had started with a short screening test aimed at estimating participant eligibility. We used illustrations of various types of physical activity and pictorial Likert-type scales to promote participants' understanding of the questions and reliability of their responses. We also adopted several strategies suggested by Finlay and Lyons (2001) for administering self-report questionnaires with individuals who have intellectual disabilities. These strategies included (a) response format: use of visual aide and use of fewer Likert-type points (i.e., 3-point Likert scale) as well as inclusion of a “don't know” option; (b) question phrasing: avoiding abstract concepts and negative wording (e.g., avoiding adding “no” and “not” to positive statements); (c) time questions: use of significant events as markers; and (d) abstract or general concepts: checking the meaning of the answer and using concrete situations or events to allow the interviewee to make generalized judgments. The participants were encouraged to respond to interview questions independently.

Analyses

We computed Cronbach's alpha coefficients (Cronbach, 1951) to assess the internal consistency of the Process of Change Scale, Self-Efficacy Scale, Decisional Balance Scale, and their corresponding subscales. There was acceptable internal consistency for a majority of the scales and their subscales, except the perceived cons, consciousness raising, environment reevaluation, and counter conditioning, α < .60 (see Table 2).

Table 2

Internal Consistency of Psychosocial Scales and Subscales

Internal Consistency of Psychosocial Scales and Subscales
Internal Consistency of Psychosocial Scales and Subscales

In the stages of change measure, the contemplation and action stages with small numbers of participants (n < 20) were combined into precontemplation and maintenance stages, respectively, to avoid adverse impact in the main statistical analyses (Tabachnick & Fidell, 2001). We calculated and used summative scores of each of the transtheoretical model psychosocial scales in the following statistical analyses.

We conducted a one-way multivariate analyses of covariance (MANCOVA) adjusted for memory and orientation scores to describe the relationships among the three stages of change (independent variable) and transtheoretical model psychosocial constructs (dependent variables: behavioral strategies, cognitive strategies, self-efficacy, perceived pros, and perceived cons). Further, we computed correlation coefficients to examine the relationship among transtheoretical model psychosocial constructs.

For the identification of the classification accuracy and the most relevant stages of change predictors among the transtheoretical model psychosocial constructs, we performed a direct discriminant function analysis using stage of change as the dependent variable and behavioral processes of change, cognitive processes of change, self-efficacy, perceived pros, and perceived cons as independent variables.

Results/Discussion

Stage Distribution

The stage distribution was as follows: (a) precontemplation, n  =  54 (44.6%), (b) contemplation, n  =  11 (9.1%), (c) preparation, n  =  21 (17.4%), (d) action, n  =  6 (5.0%), and (e) maintenance, n  =  29 (23.9%). The stage distribution showed an over-representation in the precontemplation stage (44.6%) and an under-representation in the maintenance stage (23.9%), compared to other populations (college students, women, adolescents, older adults, adults with physical disabilities) of different countries (Australia, Canada, United States of America, and United Kingdom), different settings (worksite, community, primary care, and education), different age groups, and different recruitment methods (passive or active) (precontemplation, M  =  14%, range  =  3 to 27; maintenance, M  =  36%, range  =  30 to 54) (Kosma, Gardner, Cardinal, Bauer, & McCubbin, 2006; Marshall & Biddle, 2001). The over- and under-representations of both these extreme stages suggest the possibility that a majority of the sample with mild intellectual disabilities engages in no or limited regular physical activity. Specifically, according to the data collected in the present study, it appears that a relatively small portion of Taiwanese adults with mild intellectual disabilities (28.9%) engaged in regular physical activity at moderate and vigorous intensity levels compared to what has been reported for adults with intellectual disabilities in the United States (42 to 47%: Draheim, Williams, & McCubbin, 2002) and compared to reports on the Taiwanese general population (65%: Liu, Chen, & Chen, 1998). This would suggest that the health benefit of physical activity may be less for Taiwanese adults with mild intellectual disabilities compared to their peers in the United States and the general Taiwanese population.

Given the known psychological, social, and physical benefits of regular physical activity (Berlin & Colditz, 1990; Dishman & Buckworth, 1996) during leisure time, increasing the prevalence of regular physical activity in this population may enhance their overall health and quality of life. In addition, research findings have indicated that manual workers are more likely to engage in sedentary activities during their leisure time, and these workers perceived lower health status and quality of life (Kaleta, Makowiec-Dabrowska, Dziankowska-Zaborszczyk, & Jegier, 2006). Therefore, promoting regular physical activity for the population with intellectual disabilities, especially for manual workers with intellectual disabilities, is also critical. The high prevalence of participants with mild intellectual disabilities who self-reported to be precontemplators, meaning that they had no intention or plan of beginning regular physical activity, suggests the potential importance of recognizing motivational readiness before an action-oriented intervention is introduced to this population. In most physical activity interventions to date, researchers have used an action-oriented approach, but our data suggest that an educational approach that promotes motivational readiness for the population with mild intellectual disabilities should receive further attention.

Relationships Among Stages of Change and Transtheoretical Model Psychosocial Variables

The bivariate correlation coefficients of the psychosocial variables are presented in Table 3. Significant correlations were identified among all the psychosocial measures. Although the correlation between behavioral and cognitive processes of change was high (exceeded .70), multicollinearity was not observed (tolerance value range  =  .902 to .725 >.1; variance inflation factor value range  =  1.108 to 1.380 < 10; Tabachnick & Fidell, 2001). The behavioral processes of change and cognitive processes of change were considered as two separate variables in the subsequent analyses.

Table 3

Correlation Matrix for Transtheoretical Model Psychosocial Constructs

Correlation Matrix for Transtheoretical Model Psychosocial Constructs
Correlation Matrix for Transtheoretical Model Psychosocial Constructs

We conducted a one-way MANCOVA adjusted for memory and orientation scores to describe the relationships among the stage of change (independent variable) and transtheoretical model psychosocial constructs (dependent variables: behavioral strategies, cognitive strategies, self-efficacy, perceived pros, and perceived cons). Findings indicated a significant multivariate effect between the stages of change and the dependent variables, Wilk's Λ  =  .699; F(10, 224)  =  4.397, p  =  .000 < .05, η2  =  .164. The results revealed that the behavioral processes and the cognitive processes of change, η2s  =  .240 and .236, respectively, contributed most of the variance followed by self-efficacy, η2  =  .160, and perceived pros, η2  =  .068. The results of the least significant difference post hoc contrasts revealed an overall increase across stages for all dependent measures, with the exception of perceived cons, yet no distinction was found between preparation and action/maintenance stages (see Figure 1).

Figure 1

Transtheoretical model psychosocial scores across three stages of change. TTM  =  transtheoretical model.

PC  =  Precontemplation; C  =  Contemplation; P  =  Preparation; A/M  =  Action/Maintenance. Summative scores were computed for each and the scores ranged as follows: Behavioral Processes of Change, 0 to 45; Cognitive Processes of Change, 0 to 45; Self–Efficacy, 0 to 54; Perceived Pros, 0 to 15; Perceived Cons, 0 to 15.

Figure 1

Transtheoretical model psychosocial scores across three stages of change. TTM  =  transtheoretical model.

PC  =  Precontemplation; C  =  Contemplation; P  =  Preparation; A/M  =  Action/Maintenance. Summative scores were computed for each and the scores ranged as follows: Behavioral Processes of Change, 0 to 45; Cognitive Processes of Change, 0 to 45; Self–Efficacy, 0 to 54; Perceived Pros, 0 to 15; Perceived Cons, 0 to 15.

Overall, the theorized relationship among the transtheoretical model psychosocial constructs across stages was partially supported. Behavioral processes of change and cognitive processes of change provided the most stage discrimination in support of studies among people without disabilities (Marshall & Biddle, 2001) and individuals with physical disabilities (Cardinal et al., 2004; Kosma, Cardinal, & McCubbin, 2004; Kosma, Gardner, Cardinal, Bauer, & McCubbin, 2006).

The findings also indicate that participants with mild intellectual disabilities in higher stages of change (preparation and action/maintenance stages) reported higher self-efficacy and perceived pros than those in the lower stages (precontemplation and contemplation stages), yet no significant distinctions were found in perceived cons. These results are also supported by Tung, Gillett, and Pattillo (2005), who found that among Taiwanese caregivers of patients with mental illness, the perceived cons measure was unable to differentiate stages. In two other transtheoretical model studies conducted in Asian countries, the researchers also indicated no significant stage differences in perceived cons in young Chinese (Callaghan, Eves, Norman, Chang, & Lung, 2002) and Japanese women (Wakui et al., 2002). Including the results found in this study, perceived cons do not appear to distinguish between stages of change. The validity of perceived cons developed in a Western country (the United States) may be inappropriate for application to Asian countries. Future examination of cultural relevance of the Perceived Cons subscale is needed. Consequently, utilization of the transtheoretical model for physical activity promotion in Asian populations should probably focus on perceived pros and not perceived cons.

Classification Accuracy and Psychosocial Predictors of the Stage of Change

The findings of the direct discriminant function analysis produced one significant discriminant function (composite score of the predictors), Wilks' Λ  =  .72, χ2 (15, N  =  121)  =  37.80, p  =  .000 < .05; canonical R  =  .50. The variance accounted for in the between-group variability was 88.6% for the discriminant function. The group centroids (mean discriminant scores across stages) of the first canonical discriminant function across stages were: (a) precontemplation/contemplation  =  −.510, (b) preparation  =  .327, and (c) action/maintenance  =  .751. The predictors of the canonical discriminant function distinguished precontemplation/contemplation from preparation (precontemplation/contemplation < preparation) and action/maintenance (precontemplation/contemplation < action/maintenance). No difference was found between preparation and action/maintenance. After adjusting for group sizes, we found that the highest accuracy in stage classification was action/maintenance followed by precontemplation/contemplation and preparation. Overall, the classification accuracy across stages was 56.2% (see Table 4).

Table 4

Classification Accuracy Within and Across the Stages of Change

Classification Accuracy Within and Across the Stages of Change
Classification Accuracy Within and Across the Stages of Change

The classification accuracy identified through the transtheoretical model psychosocial constructs in this study was considered high (56.2%) and fell within the range previously reported (50% to 69.6%) among individuals with and without disabilities (Cardinal et al., 2004; Cardinal, Tuominen, & Rintala, 2003; Kosma et al., 2004, 2006; Marshall & Biddle, 2001). Similar to Kosma et al.'s findings (2006), our results showed that the two extreme stages, precontemplation/contemplation (58.5%) and action/maintenance (62.9%), exhibited the highest classification accuracy.

The direct discriminant function analysis indicated that the most important predictors of the stages of change entered in the equation were the behavioral processes of change, r  =  .934, cognitive processes of change, r  =  .871, self-efficacy, r  =  .736, and perceived pros, r  =  .439. Cognitive processes of change involve one's thinking, attitudes, and awareness toward physical activity (i.e., increasing knowledge, being aware of risks, caring about consequences to others, comprehending benefits, and increasing healthy opportunities), while behavioral processes of change contain behavioral strategies that enhance action (i.e., substituting alternatives, enlisting social support, rewarding yourself, committing yourself, and reminding yourself) (Prochaska & Velicer, 1997a). Self-efficacy has been reported to be a strong stage predictor in exercise behavior and is defined as one's confidence in overcoming barriers (Bandura, 1986). Based on the results of this study, individuals with mild intellectual disabilities who are in later stages (action and maintenance) may have more knowledge, better attitudes, and higher awareness about physical activity, and, therefore, engage in more leisure-time physical activities than those in earlier stages.

In addition, people in later stages tend to use more behavioral strategies to overcome difficulties, and they also perceive higher confidence to overcome barriers. Such findings have been previously reported for adults with physical disabilities (Kosma et al., 2004, 2006) and without disabilities (Marshall & Biddle, 2001). It is interesting that the memory level measured through the Modified Fuld Object-Memory Tests (Seltzer, 1997) was significantly related to the reports of cognitive processes of change, self-efficacy, and perceived pros, suggesting that participants with mild intellectual disabilities who scored higher in the Modified Fuld Object-Memory Tests reported higher cognitive processes of change, r  =  .258, p < .001, Self-Efficacy, r  =  .184, p < .05, and Perceived Pros, r  =  .169, p < .05, than did those who scored lower. This finding implies that the application of the measures to individuals with more severe intellectual disabilities may be limited. Future research is needed to investigate the impact of cognitive levels on the reliability of psychosocial reports. In addition, other ways of measuring psychosocial aspects in individuals with more limitations in their cognitive levels need to be further explored.

Conclusion

This is the first study to examine the possibility of applying the full transtheoretical model to describe physical activity engagement of adults with mild intellectual disabilities in a Taiwanese cultural context. We followed rigorous steps in gathering self-psychosocial reports of participants with mild intellectual disabilities. The theorized relationship among the transtheoretical model psychosocial constructs across the three collapsed stages was partially supported. The behavioral processes of change, cognitive processes of change, self-efficacy, and perceived pros increased across stages; yet, a clear linear pattern of association was not observed. The lack of strong linear association among the stage of change and the transtheoretical model psychosocial constructs may be due to (a) the small numbers of participants in contemplation and action stages that led to combination of stages may hinder the examination of construct relationships between regularly active groups, although combining stages is commonly seen in transtheoretical model literature; (b) some items in transtheoretical model psychosocial measures (i.e., decisional balance) may not be culturally relevant to the population with mild intellectual disabilities in a Taiwanese cultural context; and (c) the modified Likert type scale, from the original 5- to 4-point, in the transtheoretical model psychosocial measures may have created psychometric property issues that led to the lack of strength in confirmation of five stage differences among the psychosocial measures. Future investigators may consider recruiting more participants classified in each of the five stages (at least 20 per stage) to allow closer examination of the relationships among full transtheoretical model constructs.

Overall, the results of this study show some correspondence with the theorized patterns of the transtheoretical model. These results, however, can only be applied to adults with mild intellectual disabilities in a Taiwanese culture context. Caution should be taken in generalizing the results to the same population in other countries, other Taiwanese populations, or adults with more severe intellectual disabilities. Empirical data that describe the psychosocial profile of participants with intellectual disabilities need to be further explored by integrating rigorous steps to ensure reliable and valid responses reported directly by such individuals. Psychosocial measures specifically developed for a population with intellectual disabilities and in a Taiwanese culture context should be further explored, allowing culturally relevant factors to be taken into consideration. The findings of this study may provide some helpful evidence for practitioners and policymakers on application of the transtheoretical model to health promotion interventions for populations with mild intellectual disabilities.

Acknowledgments

We gratefully acknowledge the support of the Parents' Association for Persons With Intellectual Disability, Taiwan, Republic of China; other local participating agencies; and all study participants.

References

References
American College of Sports Medicine
1990
.
Position stand: The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness in healthy adults.
Medicine and Science in Sports and Exercise
22
:
265
274
.
Bandura
,
A.
1986
.
Social foundations of thought and action
.
Englewood Cliffs, NJ
Prentice-Hall
.
Banville
,
D.
and
P.
Desrosiers
.
2000
.
Translating questionnaire and inventories using a cross-cultural translation technique.
Journal of Teaching in Physical Education
19
:
374
387
.
Benisovich
,
S.
,
J. S.
Rossi
,
G. J.
Norman
, and
C. R.
Nigg
.
1998
.
Development of a multidimensional measure of exercise self-efficacy [Abstract].
Annuals of Behavioral Medicine
20
:
S190
.
Berlin
,
J. A.
and
G. A.
Colditz
.
1990
.
A meta-analysis of physical activity in the prevention of coronary heart disease.
American Journal of Epidemiology
132
:
612
629
.
Callaghan
,
P.
,
F. F.
Eves
,
P.
Norman
,
A. M.
Chang
, and
C. Y.
Lung
.
2002
.
Applying the transtheoretical model of change to exercise in young Chinese people.
British Journal of Health Psychology
7
:
267
282
.
Cardinal
,
B. J.
,
M.
Kosma
, and
J. A.
McCubbin
.
2004
.
Factors influencing the exercise behavior of adults with physical disabilities.
Psychobiology and Behavioral Strategies
36
:
868
875
.
Cardinal
,
B. J.
,
K. J.
Tuominen
, and
P.
Rintala
.
2003
.
Psychometric assessment of Finnish versions of exercise-related measures of transtheoretical model constructs.
International Journal of Behavioral Medicine
10
:
31
44
.
Cronbach
,
L. J.
1951
.
Coefficient alpha and the internal structure of tests.
Psychometrika
16
:
135
138
.
Dishman
,
R. K.
and
J.
Buckworth
.
1996
.
Increasing physical activity: A quantitative synthesis.
Medicine and Science in Sports and Exercise
28
:
706
719
.
Draheim
,
C. C.
,
D. P.
Williams
, and
J. A.
McCubbin
.
2002
.
Prevalence of physical inactivity and recommended physical activity in community-based adults with mental retardation.
Mental Retardation
40
:
436
444
.
Dunn
,
A. L.
1996
.
Getting started: A review of physical activity adoption studies.
British Journal of Sports Medicine
30
:
193
199
.
Finlay
,
W. M.
and
E.
Lyons
.
2001
.
Methodological issues in interviewing and using self-report questionnaires with people with mental retardation.
Psychological Assessment
13
:
319
335
.
Folstein
,
M.
,
S.
Folstein
, and
P.
McHugh
.
1975
.
The mini-mental state examination.
Journal of Psychiatric Research
12
:
189
198
.
Heller
,
T.
,
K.
Hsieh
, and
J.
Rimmer
.
2004
.
Attitudinal and psychological outcomes of a fitness and health education program on adults with Down syndrome.
American Journal on Mental Retardation
109
:
175
185
.
Kaleta
,
D.
,
T.
Makowiec-Dabrowska
,
E.
Dziankowska-Zaborszczyk
, and
A.
Jegier
.
2006
.
Physical activity and self-perceived health status.
International Journal of Occupational Medicine and Environmental Health
19
:
61
69
.
Kosma
,
M.
,
B. J.
Cardinal
, and
J. A.
McCubbin
.
2004
.
Predictors of physical activity stage of change among adults with physical disabilities.
American Journal of Health Promotion
19
:
114
117
.
Kosma
,
M.
,
R. E.
Gardner
,
B.
Cardinal
,
J. J.
Bauer
, and
J. A.
McCubbin
.
2006
.
Psychosocial determinants of stages of change and physical activity among adults with physical disabilities.
Adapted Physical Activity Quarterly
23
:
49
64
.
Liu
,
Y. M.
,
L. H.
Chen
, and
J. J.
Chen
.
1998
.
Willingness in participation of fitness testing and its related factors (grant report)
.
Taipei
Mei-Tzao Clinic
.
Marcus
,
B. H.
,
T. K.
King
,
M. M.
Clark
,
B. M.
Pinto
, and
B. C.
Bock
.
1996
.
Theories and techniques for promoting physical activity behaviors.
Sports Medicine
22
:
321
331
.
Marcus
,
B. H.
and
L. R.
Simkin
.
1993
.
The stages of exercise behavior.
Journal of Sports Medicine and Physical Fitness
33
:
83
88
.
Marshall
,
S. J.
and
S. J. H.
Biddle
.
2001
.
The transtheoretical model of behavior change: A meta-analysis of applications.
Annals of Behavioral Medicine
23
:
229
246
.
Martin
,
J. E.
and
P. M.
Dubbert
.
1982
.
Exercise applications and promotion in behavioral medicine: Current status and future directions.
Journal of Consulting and Clinical Psychology
50
:
1004
1007
.
Nigg
,
C. R.
,
G. J.
Norman
,
J. S.
Rossi
, and
S. V.
Benisovich
.
1999
.
Processes of exercise behavior change: Redeveloping the scale
Poster session presented at the meeting of the Society of Behavioral Medicine, San Diego, CA.
Pate
,
R. R.
,
M.
Pratt
,
S. N.
Blair
,
W. L.
Haskell
,
C. A.
Macera
,
C.
Bouchard
, et al
.
1995
.
Physical activity and public health. A recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine.
Journal of the American Medical Association
273
:
402
407
.
Plotnikoff
,
R. C.
,
C.
Blanchard
,
S. B.
Hotz
, and
R.
Rhodes
.
2001
.
Validation of the decisional balance scales in the exercise domain from the transtheoretical model: A longitudinal test.
Measurement in Physical Education and Exercise Science
5
:
191
206
.
Prochaska
,
J. O.
and
W. F.
Velicer
.
1997a
.
The transtheoretical model of health behavior.
American Journal of Health Promotion
12
:
38
48
.
Prochaska
,
J. O.
and
W. F.
Velicer
.
1997b
.
Introduction.
American Journal of Health Promotion
12
:
6
7
.
Sallis
,
J. F.
and
M. F.
Hovell
.
1990
.
Determinants of exercise behavior.
In
Holloszy
,
J. O.
and
K. B.
Pandolf
. (
Eds.
).
Exercise and Sports Science Review
.
18
:
307
330
.
Baltimore
Williams & Wilkins
.
Sano
,
M.
,
P. S.
Aisen
,
A. J.
Dalton
,
H. P.
Andrews
, and
W. Y.
Tsai
.
2005
.
Assessment of aging individuals with Down syndrome in clinical trials: Results of baseline measures.
Journal of Policy and Practice in Intellectual Disabilities
2
:
126
138
.
Seltzer
,
G. B.
1997
.
Modified FULD object memory evaluation
Available from the Waisman Center, University of Wisconsin-Madison, Madison, WI 53705.
Tabachnick
,
B. G.
and
L. S.
Fidell
.
2001
.
Using multivariate statistics
.
(4th ed.).
Needham Heights, MA
Allyn & Bacon
.
Taylor
,
C. B.
,
J. F.
Sallis
, and
R.
Needle
.
1985
.
The relation of physical activity and exercise to mental health.
Public Health Report
100
:
195
202
.
Temple
,
V. A.
,
G. C.
Frey
, and
H. I.
Stanish
.
2006
.
Physical activity of adults with mental retardation: Review and research needs.
American Journal of Health Promotion
21
:
2
12
.
Tung
,
W. C.
,
P.
Gillett
, and
E.
Pattillo
.
2005
.
Applying the transtheoretical model to physical activity in family caregivers in Taiwan.
Public Health Nursing
22
:
299
310
.
U.S. Department of Health and Human Services
1996
.
Physical activity and health: A report of the Surgeon General
.
U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion
.
Atlanta
.
U.S. Department of Health and Human Services
2002
.
Physical activity fundamental to preventing disease
.
U.S. Department of Health and Human Services
2005
.
The Surgeon General's call to action to improve the health and wellness of persons with disabilities
.
Washington, DC
Author
.
Vallerand
,
R. J.
1989
.
Vers une methodologie de validation trans-culture de questionnaires psychologiques: Implications pour la rescherche en langue francaise [Toward a methodology of cross-culture validation of psychological questionnaire: Implications for research, in French].
Psychologie Canadienne
30
:
662
680
.
Vallerand
,
R. J.
and
W. R.
Halliwell
.
1983
.
Vers une méthodologie de validation trans-culturelle de questionnaires psychologiques: Implications pour la psychologie du sport.
Journal Canadien Des Sciences Appliquées Au Sport
8
:
9
18
.
Wakui
,
S.
,
T.
Shimomitsu
,
Y.
Odagiri
,
S.
Inoue
,
T.
Takamiya
, and
Y.
Ohya
.
2002
.
Relation of the stages of change for exercise behaviors, self-efficacy, decisional-balance, and diet-related psycho-behavioral factors in young Japanese women.
Journal of Sports Medicine and Physical Fitness
42
:
224
232
.
World Health Organization
2005
.
Physical inactivity for various population groups: Alarming global trend of physical inactivity.
.
World Health Organization/International Association for the Study of Obesity/International Obesity Task Force
2000
.
The Asia-Pacific perspective: Redefining obesity and its treatment
.
Sydney
Health Communications
.

Author notes

Sharon Hsu, PhD (E-mail: sharon.hsu@csun.edu), Assistant Professor, California State University, Northbridge, Department of Kinesiology, 18111 Nordhoff St.. Northridge, CA 91330-8287. Bo Fernhall, PhD, Professor, University of Illinois at Urbana-Champaign, Department of Kinesiology and Community Health, 227 Freer Hall, 906 S. Goodwin Ave. MC 052, Urbana, IL 61801. Stanley Sai Chuen Hui, PhD, Professor, The Chinese University of Hong Kong, Department of Sports Science and Physical Education, Shatin, Hong Kong. James Halle, PhD, Professor, University of Illinois at Urbana-Champaign, Department of Special Education, 284F Education Building, 1310 S. 6th St. MC 708, Champaign, IL 61820.

Editor-in-charge: Steven J. Taylor