| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
a Centre on Aging and Department of Sociology, University of Manitoba, Winnipeg, Canada
b Alberta Seniors, Edmonton, Alberta, Canada
c College of Human Services, Arizona State University West, Phoenix
d Wellness Institute, Seven Oaks General Hospital, Winnipeg, Manitoba, Canada
Correspondence: Laurel A. Strain, PhD, Centre on Aging, 338 Isbister Building, University of Manitoba, Winnipeg, Manitoba, Canada R3T 2N2. E-mail: laurel_strain{at}umanitoba.ca.
Decision Editor: Laurence G. Branch, PhD
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionAppendixReferences |
|---|
Key Words: Recreation • Aging • Longitudinal study
Gerontologists and leisure researchers share a long-standing interest in leisure participation in later life. Older adults have been reported to spend a great deal of time engaged in a variety of leisure activities (Verbrugge, Gruber-Baldini, and Fozard 1996
). Solitary activities such as watching television and listening to the radio have been cited as the most popular pursuits among this age group (Bevil, O'Connor, and Mattoon 1993), whereas sports and exercise programs are activities in which participation is least likely (Armstrong and Morgan 1998).
Several researchers (Freysinger, Alessio, and Mehdizadeh 1993; Iwasaki and Smale 1998; McGuire 2000; Singleton, Forbes, and Agwani 1993; Stanley and Freysinger 1995; Verbrugge et al. 1996) have called for studies of age-related changes in leisure activities among older adults. Mannell 1999 and Lefrancois, Leclerc, and Poulin 1998 argue for consideration of age in conjunction with other characteristics. Continuity theory (Atchley 1993) suggests that changes in activity are a function of adaptations made in response to changing life circumstances. In Baltes and Baltes 1990 model of selective optimization with compensation, individuals are thought to reduce engagements based on declining energy levels and thus selectively invest in specific pursuits. Similarly, Iso-Ahola 1980 has postulated that leisure engagement is based on an individual's desire to seek familiar or novel forms of activity and that, in later life, familiarity is more important than novelty.
Comparisons across existing studies of changes in leisure participation are difficult, given variations in definition and measurement of leisure participation, time periods, age groups, and analytical approaches. Cross-sectional studies have relied on retrospective reporting of activities ceased and/or added in a specific time period (McGuire, O'Leary, Yeh, and Dottavio 1989; Searle, Mactavish, and Brayley 1993; Zimmer, Hickey, and Searle 1997). In longitudinal studies, researchers have examined changes in overall scores for frequency of participation across a number of different activities (Iwasaki and Smale 1998; Stanley and Freysinger 1995), changes in the number of leisure pursuits (Iwasaki and Smale 1998), changes in the estimated time spent on specific activities (Armstrong and Morgan 1998; Bennett 1998; Iwasaki and Smale 1998), and the number of years over a specified period that respondents report either adding or ceasing an activity (Singleton et al. 1993). Outdoor activities (McGuire et al. 1989), self-identified activities (Searle et al. 1993; Zimmer et al. 1997), and researcher-identified activities (Iwasaki and Smale 1998) have been studied. The time periods range from 6 months to 16 years. Respondents' ages vary, with samples aged 20 and older (e.g., Iwasaki and Smale 1998; Verbrugge et al. 1996), 50 and older (Freysinger et al. 1993), and 65 and older (Armstrong and Morgan 1998; Bennett 1998). Studies conducted on older adults specifically are rare.
The limited existing longitudinal research sheds some light on changes in leisure participation as individuals age. Armstrong and Morgan 1998 and Bennett 1998 found that there tends to be a gradual decrease in the amount of time spent in leisure activities as one ages, consistent with Verbrugge and colleagues 1996 findings. The frequency of overall leisure participation has been reported to decrease over time for both older married men and women (Freysinger et al. 1993). Stanley and Freysinger 1995 found that men had significant decreases in frequency of participation in six activity categories, whereas there were significant declines for women in only two categories. In Singleton and colleagues 1993 study, men with higher incomes and more education were likely to have more years during which activities were added or ceased than men with lower incomes and less education. Retirement, especially for men, has been noted to be related to an increased frequency of leisure activity participation (Iwasaki and Smale 1998). Widowhood is another life event that may affect leisure.
Health status is also important when examining changes in leisure participation (Mannell 1999). In a cross-sectional study of individuals aged 55 and older (Searle and Iso-Ahola 1988), better self-assessed health was associated with higher rates of participation and participation in a greater number of activities. In terms of specific activities, health status was related to participation in activities that tended to be more physical in nature such as sports, physical exercise, camping, gardening, and walking, as well as volunteer work, taking lessons or formal instruction, visiting libraries/museums/art galleries, and watching live events.
Overall, the extent to which older adults continue or cease their leisure participation and characteristics associated with these changes warrants further attention. In particular, longitudinal research is required on changes in both overall and specific leisure activities and the effect of sociodemographic and health characteristics such as age, gender, marital status, education, employment status, self-rated health, and functional ability. The remainder of this article provides such an empirical examination. Of particular interest is the extent to which older adults continue their leisure participation overall and in specific activities over an 8-year period, characteristics associated with continuing versus ceasing activities, and the variations in these characteristics according to the activity under consideration.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionAppendixReferences |
|---|
; Strain 1991). The sample was drawn from the Manitoba Health Services Commission database considered to be the most complete listing available because all residents of this Canadian province must be registered to receive coverage of medical services.
In 1993, 390 of the original 743 individuals were re-interviewed (see Sweiden and Strain 1995 for details). Since 1985, 232 respondents had died, 26 were ineligible because of health (n = 25) or language problems (n = 1), 23 could not be located, 21 had moved to personal care homes, 15 had moved to other communities, and three interviews were incomplete. Thirty-three refused a second interview, yielding a refusal rate of 8%. The personal interviews lasted on average 105 min and included questions on sociodemographic characteristics, health status, and participation in leisure activities. A total of 380 respondents answered the questions on leisure activity participation in both 1985 and 1993 and are included in these analyses.
Measures
Leisure Activity Participation.
In 1985, respondents reported on the frequency (never, sometimes, often) of their participation in a series of specific leisure activities. In 1993, individuals were asked if they were able to and did participate in these same activities and if so, their frequency of participation (rarely, sometimes, often). The activities were dining out, outdoor yardwork, playing cards, reading, shopping, theater/movies/spectator sports, travel, walking, watching television, and church services/activities.
For each activity, we considered individuals who indicated that they never or rarely participated as nonparticipants (0 = nonparticipant) and we classified those who participated sometimes or often as participants (1 = participant). We calculated the number of activities that respondents participated in at least sometimes in 1985 and in 1993, with possible scores ranging from 0 to 10 in each time period.
To examine changes in participation for each of the 10 activities, we created a dichotomous variable of ceased versus continued participation (0 = ceased [participated in 1985 only], 1 = continued [participated in 1985 and 1993]). For changes in overall leisure participation, the number of activities in which participation was reported in both 1985 and 1993 was determined. Possible scores ranged from 0 to 10, indicating the number of activities that each respondent continued.
Sociodemographic and Health Characteristics.
We examined selected 1985 sociodemographic and health characteristics and changes in these characteristics between 1985 and 1993. The sociodemographic characteristics included age (continuous), gender (1 = female), marital status (1 = married), a change in marital status (1 = loss of partner), years of education (continuous), employment status (1 = employed), and a change in employment status (1 = employed in 1985 and not employed in 1993).
In terms of health, we included self-rated health and limitations in basic and instrumental activities of daily living (ADLs; IADLs) in the analysis. In both 1985 and 1993, respondents were asked: "For your age, would you say, in general, your health is excellent, good, fair, poor, or bad?" We recoded the 5-point scale into two categories (0 = excellent/good, 1 = fair/poor/bad) because of the small number of cases in some categories. We calculated a change in self-rated health from 1985 to 1993 (0 = better rating/no change, 1 = poorer rating [e.g., good in 1985 and poor in 1993]) based on the 5-point scale. We examined the ability to perform 12 ADLs/IADLs (bathing, doing household tasks, dressing, eating, handling money, preparing meals, shopping, taking medication, taking out trash, toileting, using the telephone, walking). Respondents rated their ability to perform each activity on a scale of 1 to 4 (1 = without help, 2 = with some help, 3 = only with help, 4 = unable to do). We summed responses to obtain the number of activities with which individuals required at least some assistance (Cronbach's 
= .79). We recoded the 1985 measure into no ADL/IADL limitations (coded 0) and at least one limitation (coded 1). We calculated a change in ADL/IADL limitations (0 = fewer/same number of limitations in 1993 than in 1985, 1 = more limitations in 1993 than in 1985) by subtracting the total number of ADLs/IADLs in 1993 from the number in 1985.
Data Analyses
We began the analysis with descriptive statistics of overall and specific activity participation in 1985 and changes over the 8-year period. We then turned our attention to the relationship of change in overall and specific activities with 1985 characteristics and changes in these characteristics. We conducted analysis of variance, cross-tabulations, and t tests to examine bivariate relationships. Multivariate analyses included linear regressions when the dependent variable was the total number of activities that were continued and logistic regressions when the dependent variable was cessation versus continuation of a specific activity. We developed two regression models for each dependent variable. Model 1 included 1985 sociodemographic and health characteristics; Model 2 included the same 1985 characteristics and the changes in characteristics by 1993. Checks for multicollinearity revealed difficulty with employment status and change in employment status which were highly correlated (.91). As neither variable emerged as statistically significant in bivariate analyses, we excluded both from the regression analyses.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionAppendixReferences |
|---|
Between 1985 and 1993, 16% lost a marital partner and 19% were no longer employed. Twenty-six percent had poorer self-rated health in 1993 than in 1985, and 28% had more ADL/IADL limitations in 1993 than in 1985.
Overall and Specific Leisure Activity Participation
In 1985, 68% of the respondents reported participation in at least 8 of the possible 10 activities. The number of activities ranged from 3 to 10 (Median = 8.0, M = 8.12, SD = 1.49). Respondents were most likely to participate in watching television, reading, and shopping in 1985 (Table 1 ). Activities least likely to be reported were theater/movies/spectator sports, playing cards, and church services/activities.
|
Characteristics Associated With Leisure Activity Participation
We examined characteristics associated with changes in the number of activities and in specific activities next. We excluded watching television and reading from the analyses of specific activities as over 90% continued those activities between 1985 and 1993.
Number of Activities Continued.
In Model 1, age, education, self-rated health, and ADL/IADL limitations emerged as significant (Table 2 ). Being younger, having more education, rating their own health as excellent/good, and having no ADL/IADL limitations in 1985 were associated with a greater likelihood of continuing more activities.
|
Shopping.
In regards to specific activities, 83% of the 354 individuals who shopped in 1985 continued to do so in 1993. In Model 1, age was the only significant factor, with younger respondents more likely to continue shopping (Table 3 ). In Model 2, age and a change in ADL/IADL limitations were significant. Individuals who had fewer or the same number of ADL/IADL limitations in 1993 than in 1985 were more likely to continue the activity.
|
Walking.
Seventy-nine percent of 340 individuals participated in walking in both 1985 and 1993. In Model 1, age was associated with continuing to walk, with younger respondents more likely to do so (Table 3 ). In Model 2, a change in self-rated health was the only significant factor. Having better or the same self-rated health was associated with continued walking.
Travel.
Sixty-seven percent of the 316 individuals who travelled in 1985 continued to do so in 1993. In Models 1 and 2, age and self-rated health were significant (Table 3 ). Individuals who continued the activity tended to be younger and rated their health as excellent or good in 1985.
Outdoor Yardwork.
Of the 271 individuals who participated in outdoor yardwork in 1985, 68% continued their participation in 1993. In Model 1, age and self-rated health were significant once again (Table 3 ). Being younger and having excellent or good self-rated health was associated with continuing outdoor yardwork. In Model 2, self-rated health, a change in marital status, and a change in ADL/IADL limitations were significant. Age was not significant once we took into account change characteristics. Not losing a marital partner over the 8-year period and having fewer or the same number of ADL/IADL limitations in 1993 than in 1985 were characteristics associated with a greater likelihood of continuing outdoor yardwork.
Church Services/Activities.
Seventy-three percent of the 251 participants in church services/activities in 1985 continued this activity in 1993. In Model 1, age was statistically significant (Table 3 ). Younger individuals were more likely to continue with church services/activities. In Model 2, gender and a change in ADLs/IADLs were significant whereas age was not. Women and individuals who had fewer or the same number of ADL/IADL limitations in 1993 than in 1985 were more likely to continue this activity.
Playing Cards.
Of the 246 individuals who played cards in 1985, 67% continued playing in 1993. None of the characteristics we considered were significantly associated with continuation of this activity (Table 3 ).
Theater/Movies/Spectator Sports.
Almost one half (45%) of the 224 participants in 1985 continued to attend theater/movies/spectator sports in 1993. In Model 1, age, gender, and education were significant (Table 3 ). Being younger, being female, and having more education were related to continued participation. In Model 2, these three characteristics were again significant as was a change in self-rated health. Having better or the same self-rated health in 1993 than in 1985 was associated with continuing this activity.
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionAppendixReferences |
|---|
; Bennett 1998). Yet age alone does not explain why some older adults cease their participation nor is it significantly associated with all activities studied here.
The importance of examining factors in addition to age is readily apparent. Self-rated health and functional ability are significantly related to changes in activities albeit with no consistent pattern. Of particular interest is the finding that a change in functional ability, rather than functioning at baseline, is related to change in some specific activities. This suggests a compatibility between functional ability and leisure participation that shifts as functioning changes. It highlights the possibility that individuals reduce participation in a planned manner in response to circumstances, thereby supporting the adaptation argument in continuity theory and Baltes and Baltes 1990 notion of selective investment of energy.
Life events or transitions also appear to have an impact on leisure participation to some extent. Previous research (Chiriboga and Pierce 1993) has argued that changes over time may be a function of life events or life transitions such as loss of a spouse, physical relocation, or other similar types of stressful factors. In this study, losing a partner during the 8-year period was associated with continued participation in fewer activities overall, and ceasing outdoor yardwork in particular. This suggests that change in participation is not always contingent on losing a marital partner with whom one can participate. Information on the relationship of activity partners might help to identify activities most likely to be affected by a change in marital status. There was no effect of retirement on changes in leisure participation, in part due to the small numbers who experienced this life event.
Education and gender play a relatively minor role in relation to changes in leisure activities. Having more education is associated with continuing a greater number of activities, and specifically with dining out and attending theater/movies/spectator sports. Education may serve as a proxy for income and thus is related to the activities that are more likely to incur costs. Pohjolainen 1991 has suggested that education influences the life-style choices of older adults. Women are more likely to continue their participation in church services/activities and theater/movies/spectator sports, possibly because of a greater likelihood to seek out these types of social settings than men. The lack of gender differences may reflect the activities we considered, although Stanley and Freysinger 1995 and Verbrugge and colleagues 1996 have previously reported that women's activity levels remained more constant over time than men's did.
Several issues arise that require further study. The relationship between age and changes in leisure participation warrants more in depth attention. Researchers need to explicitly test whether age is a proxy for understanding the impact of life events and changes in health on leisure participation or whether there is greater resilience at a younger age that results in higher level abilities to adapt. Details about changes in sociodemographic and health characteristics were not available in the current study. Information such as when a change occurred might shed further light on changes in participation in leisure activities. Characteristics such as nonleisure demands on one's time also need to be studied.
Changes in leisure participation such as continuing but reducing the frequency or intensity of participation, adding new activities, or replacing activities require attention. The current study focused only on continuing or ceasing activities. We did not examine the adding of activities specifically, although it is evident that relatively few respondents added any of the listed activities to their leisure repertoire. In addition, information was collected at two points over an 8-year time period and thus changes during that time period are not known.
We considered a limited number of leisure activities here and although these activities are frequently discussed in the literature, they may not be most important to older adults. Related to this, the distinction between leisure activities versus ADLs and IADLs that indicate functional ability needs to be further explored. In this study, we considered shopping and walking as leisure activities that were also 2 of the 12 activities used to determine ADL/IADL limitations (see A Note). The extent to which respondents viewed the activities as leisure or everyday tasks is not known.
Clearly there is a need to better understand the value older individuals themselves place on leisure activities, the importance of familiar versus novel activities (Iso-Ahola 1980), and possible changes in such values in the context of reductions of physical functions, life events, and life transitions. According to Baltes and Baltes 1990, individuals make conscious choices to reduce their activities irrespective of domain (e.g., work, housework, leisure, caregiving, etc.) to conserve their energy for those highly valued. Older adults' own explanations of changes in their leisure participation will shed light on this decision-making process.
Individuals who decline in their participation, whether through conscious decision-making or as a result of a traumatic life event or transition, could possibly benefit by reviewing their situation and examining whether their decisions are truly optimal for their circumstances. Research by Searle and colleagues (Searle, Mahon, Iso-Ahola, Sdrolias, and van Dyck 1995, Searle, Mahon, Iso-Ahola, Sdrolias, and van Dyck 1998) suggests that older adults with functional limitations who withdraw from desired leisure activities respond well to a leisure education intervention. These researchers reported short-term gains in leisure control, leisure competence, and reduced boredom for the experimental group 5 months following the intervention. Helping older adults address constraints that mitigate leisure choices or preferences, and their successful engagement in social, physical, intellectual, or other pursuits can be instrumental in sustaining healthy lifestyles, independence, and well-being.
The Forum
Book Reviews
Practice Concepts
|
Acknowledgments |
|---|
Received for publication August 8, 2001. Accepted for publication November 15, 2001.
|
Appendix |
|---|
|
TOPAbstractMethodsResultsDiscussionAppendixReferences |
|---|
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionAppendixReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  G. T. Nguyen, M. N. Wittink, G. F. Murray, and F. K. Barg More Than Just a Communication Medium: What Older Adults Say About Television and Depression Gerontologist, June 1, 2008; 48(3): 300 - 310. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Nilsson, B. Lofgren, A. G. Fisher, and B. Bernspang Focus on Leisure Repertoire in the Oldest Old: The Umea 85+ Study Journal of Applied Gerontology, November 1, 2006; 25(5): 391 - 405. [Abstract] [PDF]
|
|
|
|
 |
|
 N. Agahi, K. Ahacic, and M. G. Parker Continuity of leisure participation from middle age to old age. J. Gerontol. B. Psychol. Sci. Soc. Sci., November 1, 2006; 61(6): S340 - S346. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. H. Menec The Relation Between Everyday Activities and Successful Aging: A 6-Year Longitudinal Study J. Gerontol. B. Psychol. Sci. Soc. Sci., March 1, 2003; 58(2): S74 - 82. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
