Introduction
Korean society is characterised as having undergone a compression of modernity, which refers to a phenomenon involving condensed social changes that have rapidly transformed Korea into a modern society (Chang Reference Chang2003). It has experienced Japanese colonisation, the Korean War, expedited economic development and democratisation over the last half century. Moreover, contemporary Korean society is experiencing an exceptionally rapid demographic shift to an ageing society due to a low fertility rate and the increasing number as well as proportion of the older adult population (Eun Reference Eun2007; Phang et al. Reference Phang, Shin, Kim and Shin2004). This rapid social and demographic change has influenced the traditional institution of the family and has resulted in changes to the status and role of the older population in Korea (Chung Reference Chung2004; Kim Reference Kim2009). Changes from traditional extended family networks to nuclear family structures have caused an erosion of the filial piety principal that was formerly the core of traditional Korean culture and which gave older adults authority in decision making and the transfer of economic resources to their children (Kim Reference Kim2002; Nam Reference Nam2004; Park Reference Park2003). Changing traditional institutions are not only associated with a decrease in the status of older adults in the family. They have also led to a public perception of older adults as an unproductive burden to the family and society in an ageing society.
Age not only indicates chronological years since birth, but it also carries implicit and explicit social meanings. Expectations regarding age-appropriate behaviours constitute age norms that dictate behaviours, roles, timing of life events and social interactions (Elder Reference Elder1975; Settersten Reference Settersten, Mortimer and Shanahan2003; Settersten and Mayer Reference Settersten and Mayer1997). The role of an older adult in the traditional extended family as the family head is not maintained in an urbanised society and older adults often experience confusion regarding their socially expected roles and behaviours when they do not have prescribed social roles (Yang Reference Yang2004, Reference Yang2012). Age norms, defined as shared expectations regarding age-appropriate behaviours, are perceived differently depending on the age group to which individuals belong. The nuclear family structure has limited formal and informal opportunities for intergenerational contact and these changes, combined with an increasing gap in lifestyles and values between young and older generations, are considered to have produced biased images of older adults. Generational conflict, as a new structure of social and political conflict in ageing Korean society, is mainly framed within a generational equity debate defined as competition over employment, welfare spending and long-term care costs (Park Reference Park2011), as well as growing tension between younger and older generations regarding elder care in the family (Lee, Choi and Kim Reference Lee, Choi and Kim2010). These aspects of generational conflict may appear differently across age groups. Competition over employment could be a more significant source of conflict between younger and older people, while the economic and psychological burden of elder care in the family might be a more significant issue between middle-aged and older people. On the other hand, older Koreans tend to maintain traditional values of Korean culture such as filial piety and familism often demonstrated by an emphasis on the collective concerns of the family over individual family members (Paik and Choi Reference Paik and Choi2007). While studies have reported that images of older adults in Korea tend to be negative (Kim Reference Kim2002; Won Reference Won2004) and conflictive perspectives in viewing intergenerational relations have intensified, there is a lack of empirical studies on how negative images of older adults and attitudes towards generational relations might influence actual perceptions about appropriate behaviours and the social roles of older adults in contemporary Korean society.
In this paper, we explore age norms approached as shared expectations of appropriate behaviours for older adults in Korean society. Using the ageing theories of disengagement, modernisation and activity, we focus on examining how much perceptional restrictions Koreans would place on the social participation and engagement of older adults in various behaviours. In doing so, we focus on exploring whether prejudice towards older adults and generational conflict assessed as the degree to which people hold competitive attitudes in intergenerational relationships and perceive values of intergenerational integration in dealing with issues in an ageing society, are associated with attitudes towards appropriate behaviours of older adults. Further, we are interested in examining if these associations vary across generations. Studying Koreans’ expectation regarding older adults’ acceptable behaviour in the context of generational differences will help develop policies and programmes that target intergenerational integration.
Age-appropriate behaviour as an aspect of age norm
Settersten (Reference Settersten, Mortimer and Shanahan2003) has suggested that age norms can be approached through different aspects. ‘Statistical age norms’ refer to the idea that there are demographical trends that indicate the ages at which people most commonly experience major life events, such as the age at which to start formal schooling and exit the labour force. This statistical regularity in the timing of life transitions may imply the existence of a formal age structure in society that influences opportunities and constraints for different age groups (Chung Reference Chung2004; Settersten Reference Settersten, Mortimer and Shanahan2003). A second approach, described as ‘optimal age norms’, examines if there are collective notions about timetables for the transitions of life events (Byrd and Bruess Reference Byrd and Bruess1992; Gee Reference Gee1990; Neugarten, Moor and Lowe Reference Neugarten, Moore and Lowe1965; Peterson Reference Peterson1996; Settersten and Mayer Reference Settersten and Mayer1997). Optimal age norms also assess how much consensus exists among people in reporting the ‘best’ ages, which represent a shared perception of a social clock (Byrd and Bruess Reference Byrd and Bruess1992; Peterson Reference Peterson1996). A third approach focuses on the social control function of age norms that governs the timing of the engagement in certain behaviours and roles (Marini Reference Marini1984; Rook, Catalano and Dooley Reference Rook, Catalano and Dooley1989; Settersten Reference Settersten, Mortimer and Shanahan2003). Studies using age norms from this perspective have focused on the perceived consequences of satisfying or failing to satisfy generally shared timetables of life transitions (Settersten and Hagestad Reference Settersten and Hagestad1996a, Reference Settersten and Hagestad1996b).
Other scholars have considered age norms as perceptions which satisfy the rules underlying appropriate behaviour for certain ages or life stages. For example, Hori (Reference Hori1994) has suggested that there are timing, desirability and intensity aspects to age norms. His conceptualisation of timing is similar to the ‘optimal age norm’, as it suggests that people in society believe there is a perceived schedule for the transition of life events. The desirability aspect indicates evaluative criteria to determine if certain behaviours are desirable or meet age-appropriate expectations and the intensity of age norms represents the degree to which age norms are believed in society. While the studies discussed so far have assessed age norms as social rules encompassing the adult lifespan, Lee's (Reference Lee2006) study assessed age norms as shared perceptions of the age-appropriate behaviours of older adults based on a sample of older Koreans. Using activity and disengagement theories of ageing, Lee examined whether older Koreans were more likely to conform to behavioural norms that were oriented from either activity or disengagement theory in various aspects of their lives, including work, family, marriage and gender roles, intergenerational relationships, and social activities. Lee's approach is close to the desirability aspect that Hori suggested in that he assessed the degree of behavioural constraint placed on Koreans in a variety of behaviours and activities. Although his research is pioneering in investigating age norms in Korean society, his analysis was confined to a sample of older Koreans and thus the research was limited in terms of understanding the expectations relating to the behaviour of older adults in different age groups.
From a broader perspective regarding the social function of age, we conceptualise age as a determinant of expectations on behaviours and roles in society. Based on such a conceptualisation, we examine how much age-structuring exists in determining the appropriateness of older adults' behaviours among Koreans. In this perspective, our use of age norms deviates from previous studies that have approached these norms as a perceived social clock in life transitions or as a form of social control reinforced by sanctions; rather it assesses the degree of approval and disapproval about social participation and engagement in certain behaviours in later life and how strongly such perceptions are believed among Koreans.
Theoretical perspectives on age norms for older adults
The concept of age norms that is used in this paper relates to shared perceptions of age-appropriate behaviours for older adults and it is based on the theoretical approach of disengagement theory, activity theory and modernisation theory. These theories describe processes of social ageing and how social change influences the status and roles in a modernised society. Disengagement theory (Cumming and Henry Reference Cumming and Henry1961) proposes that the process of ageing is characterised by social and psychological disengagement and that this process involves a mutual withdrawal between society and the older person. Older adults are likely to experience greater life satisfaction when they reduce previously engaged activities, and society also benefits from the process, thus encouraging disengagement through measures such as institutionalised retirement. Although disengagement theory has been criticised for not representing the actual behaviour of many older adults, recent literature argues that disengagement theory may still be relevant to explain certain aspects of later life (Achenbaum and Bengtson Reference Achenbaum and Bengtson1994; Tornstam Reference Tornstam1989).
Modernisation theory (Cowgill Reference Cowgill, Cowgill and Holms1972) has been widely used as a framework that explains the changing status of older adults in a modern society that is associated with a weakening of the filial obligation principle within family (Chee Reference Chee2000; Choi Reference Choi1996; Silverstein et al. Reference Silverstein, Burholt, Wenger and Bengtson1998). The theory posits that modernisation characterised by economic development, the expansion of health technologies and mass education, together with urbanisation, contribute to the lowering of status of older members of society (Aboderine Reference Aboderine2004). On the one hand, the process of modernisation is considered not only to be the cause of a decline in the status of older adults in Korea, but also to describe how the lowered status of older adults has led to the emergence of issues such as economic dependence, role ambiguity or role loss in family, and social and psychological alienation (Choi Reference Choi1996). On the other hand, studies have emphasised the multi-faceted nature of the elder respect value (Sung Reference Sung1995) and the practice of elder respect and filial piety in Korean families which is still present in varying and complex forms (Park et al. Reference Park, Phua, McNally and Sun2005; Sung Reference Sung1998; Sung and Song Reference Sung and Song2001) instead of the ‘monotonic weakening’ (Silverstein et al. Reference Silverstein, Burholt, Wenger and Bengtson1998: 389) of filial obligation in modernised society that the theory originally proposed.
Activity theory (Lemon, Bengtson and Peterson Reference Lemon, Bengtson and Peterson1972) posits that maintained social participation from middle-age into old-age and the engagement in new roles in later life are associated with ageing well. Such ‘busyness’ in later life has been described as a prescription for the ideal ageing experience (Katz Reference Katz2000). Activity theory has been adopted in more recent theoretical models including successful ageing and productive ageing (Sherradan et al. Reference Sherradan, Morrow-Howell, Hinterlong, Rozario, Morrow-Howell, Hinterlong and Sherradan2001). The productive ageing perspective with its focus on the recognition of the productivity of later life and the encouragement of engagement in productive activities (e.g. working, volunteering and care-giving) has gained attention from policy makers and it has been been adopted as the basis of policies concerned with population ageing in Korea (Government of the Republic of Korea 2010). In particular, an emphasis has been placed on the older population's continued participation in the labour force to achieve a sustainable national economy and productivity in the face of the massive retirement of the baby-boomer generation and resulting increases in welfare costs.
While these theories deal with different expectations regarding the ageing process, the stance of this paper is that the expectations and propositions emanating from these theories can be considered to be a continuum with some theoretical perspectives supporting disengagement whilst others emphasise continued social participation. Based on this approach, our measure of age norms can be described as the age-normative expectation of social participation and engagement in certain behaviours for older adults.
Factors affecting age norms of older adults
Individuals may show varying degrees of perceptions about age-appropriate behaviours for older adults depending on their characteristics (Hori Reference Hori1994; Lee Reference Lee2006; Peterson Reference Peterson1996; Settersten and Hagestad Reference Settersten and Hagestad1996a, Reference Settersten and Hagestad1996b). Young and older people may perceive life schedules differently, and younger people tend to believe that engaging in certain activities, for example marriage and education, is not expected to occur in older ages (Hori Reference Hori1994). Women perceive age deadlines for career trajectories as more flexible (Settersten and Hagestad Reference Settersten and Hagestad1996b), while they also have shown a greater concentration on family-related life event transitions than men (Settersten and Hagestad Reference Settersten and Hagestad1996a). Also, socio-economic status influences age norms and those with a lower education and occupational status may consider earlier ages to be appropriate for major life events than their higher education counterparts.
In addition to these demographic and socio-economic factors, perceptions on age-appropriate behaviour for older adults may be formed by attitudes toward older members of society and perceptions about different generations that co-exist in society. While age-related attitudes are theorised to exist for both young and old people and can be positive and negative, past research has asserted that negative prejudices against older adults are more pervasive (Cuddy, Norton and Fiske Reference Cuddy, Norton and Fiske2005; Hummert Reference Hummert1990; Kite et al. Reference Kite, Stockdale, Whitley and Johnson2005; Schmidt and Boland Reference Schmidt and Boland1986). Prejudice against older adults is a form of ageism, which refers to a cognitively biased evaluation toward older adults involving stereotyping, over-simplification and over-generalisation (Palmore Reference Palmore1990; Won Reference Won2004). Based on their meta-analysis of studies on age-related attitudes, Kite et al. (Reference Kite, Stockdale, Whitley and Johnson2005) have suggested that age-related attitudes involve evaluations of competence, stereotypic beliefs and perceptions of attractiveness, with older adults being perceived to be less attractive and less competent than younger people. In Korean society, negative attitudes toward older adults have been heightened by the declining social status of older adults caused by rapid industrialisation and the collapse of the principle of respect for elders (Kim Reference Kim2009; Lee, Choi and Kim Reference Lee, Choi and Kim2010; Nam Reference Nam2004). Previous studies have reported that older adults in Korea are perceived to be dependent on other people, frail, incapable of problem solving and indifferent towards sexual relations (Han Reference Han2000; Park and Yi Reference Park and Yi2001; Won Reference Won2004), and these negative images and beliefs may be more pronounced than those observed in Western societies (Han Reference Han2000).
Generational conflict in ageing societies is generally described as competition between young and older generations over limited resources, such as employment and pension benefits, alongside age-related inequality and discrimination (Park Reference Park2011; Turner Reference Turner1998). The intergenerational inequity debate, which refers to the unequal cost and benefit between the working-age population and retired older adults in receiving state-based pensions, is thought to be at the core of the generation conflict discourse (Foner Reference Foner2000). In Korea, previous research on intergenerational conflict has been framed as tension among family members resulting from the changing values related to care for older parents (Kim Reference Kim2006; Kwon Reference Kwon2004; Nam Reference Nam2004). Recently, understanding of intergenerational conflict has been expanded to indicate tensions between different generations from a broader perspective relating to population ageing and concerns about distributional justice between young and older generations that has risen in the face of the dramatically increasing costs of public pension and health-care programmes for older adults (Lee, Choi and Kim Reference Lee, Choi and Kim2010; Park Reference Park2011). In this situation of increasing generational conflict, coupled with widening gaps in the values and lifestyles between generations, political and academic discussions have called for generation integration approaches in policies related to population ageing (Chung Reference Chung2004). We conceptualise the generational conflictive perspective as maintaining competitive views between generations in relation to welfare benefits and employment and having low values for generational integration when referring to issues posed by an ageing society.
Despite the demographic shift to an ageing society and the increased attention it raises, the actual perception of behaviours of older adults by members of society remains largely under investigated in Korea. Informed by ageing theories of activity, disengagement and modernisation, we approach age norms as the degree to which people believe older people in society should or should not continue social participation and engagement in different areas of life. Based on such a conceptualisation, we examine whether negative stereotypes of older adults and perceptions about intergenerational relationships are associated with behavioural restrictions placed on older adults and whether these associations vary by different age groups.
Methods
Data and sample
Data are from the Study on Aging in Korean Society (Chung et al. Reference Chung, Choi, Bae and Lee2011), the first national-level study conducted in order to assess the attitudes of Koreans on various issues relating to population ageing, including age norms, age discrimination, generational relations and policies for older adults. Respondents aged 20 years and above were recruited through a multi-stage quota sampling technique based on administrative districts, age and gender. At the first stage, the largest seven Korean metropolitan cities, including Seoul, and nine provinces were chosen. At the second stage, these 16 cities were sorted by population size, and within these 16 cities, 40 smaller administrative districts were randomly chosen. At the final stage, a total of 1,500 respondents (750 for each gender and 500 for each of three age groups of 20–44, 45–64 and older than 64) were recruited. Face-to-face interviews were conducted in the homes of respondents by trained interviewers. From the recruited 1,500 individuals, the current analysis is based on those respondents who had information on all of the study variables, yielding a final sample size of 1,445.
Measures
Age norms for older adults
A modified age norm scale that was developed by Lee (Reference Lee2006) for older adults was utilised. The measure of age norms in the present study measures the degree to which respondents place restrictions on social participation and the engagement in various behaviours for older adults. Questions are based on either activity theory that favours engagement and social participation or disengagement and modernisation theories that suggest withdrawal as the ideal ageing process. For ten items that deal with areas including employment, education, leisure and social participation, and marriage in later life, respondents were asked to indicate the extent to which they agreed or disagreed (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5=strongly agree) with statements describing age-appropriate behaviours for older adults. Examples of the questions are: ‘People should retire in old age’, ‘Education is for younger people’, ‘It is important to enjoy leisure and social activities after retirement’ (reverse coded), ‘It is not appropriate to divorce in old age’ and ‘It is not appropriate for an older widowed individual to go on a date’ (see Appendix for all the questions). An age norm older adults scale was constructed by calculating the mean of these ten items, with higher scores indicating more restrictive attitudes in determining acceptable behaviours for older adults. The reliability score assessed by Cronbach's alpha for the scale was 0.60.
Prejudice against older adults
Previous literature has suggested that negative attitudes towards older people can include biased evaluations represented by over-generalisation or over-simplification (Palmore Reference Palmore1990; Won Reference Won2004). We employed a previously used scale for prejudice against older adults developed by Won (Reference Won2004). The 16-item scale in the present analysis measures the degree that respondents' over-simplified and negatively evaluated older adults on issues concerned with dependency, relations with other people and the potential to contribute to society. Examples of the items are: ‘Older adults tend to become dependent on other people’ and ‘There are not many ways in which older adults can contribute to society’. For each item, respondents were asked to indicate how much they agreed or disagreed (1 = strongly disagree, 2 = disagree, 3 = agree, 4 = strongly agree) with a statement reflecting prejudiced views against older adults and old age. A scale for these attitudes towards older adults was created by calculating the mean of the 16 items; higher scores indicate greater negative evaluation of older adults. Cronbach's alpha value for the scale was 0.78.
Generational conflict
We conceptualise the generational conflictive perspective as maintaining competitive attitudes between generations and having low values or perceptions on the needs of generational integration in dealing with issues in an ageing society. A modified scale of generational conflict developed by Lee and colleagues (Lee, Choi and Kim Reference Lee, Choi and Kim2010) was utilised in the analysis. The measure assesses respondents' support of competitive structures between generations in welfare benefit and employment, and perceptions on the need for generational integration when dealing with issues in an ageing society such as ideological gaps between young and older people. Respondents were asked to indicate the extent to which they agree or disagree with nine statements describing policy approaches that support generational competitive perspectives and including employment policies, the age to receive retirement pension benefits and support for intergenerational programmes (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree). Specific examples of the items include: ‘It is more important to solve issues with unemployment among young people than to create jobs for older adults in economic recession’, ‘Old age pension benefit rates should be decreased to reduce economic pressure on younger people’, ‘There should be more formal opportunities for different generations to share their own knowledge, experience, and skills’ (reverse coded). Higher scores indicate a stronger inclination towards a competitive perspective between the generations and lesser perceived needs for an intergenerational approach in dealing with issues raised by an ageing society. Reliability of the scale assessed with Cronbach's alpha value for the scale was 0.68.
Covariates
The analyses also included a set of covariates known to influence age norms, prejudice against older adults and generational conflict. These covariates include age (0 = 20–44, 1 = 45–64, 2 = 65+), gender (1 = female), marital status (1 = married), employment status (1 = working), educational attainment (0 = middle school graduate or less, 1 = high school graduate, 2 = college graduate or more), household income (in Korean Won), number of social meetings that the respondent attends regularly (0 = none, 1 = one, 2 = more than two) and self-rated health status (0 = poor or fair, 1 = good, 2 = excellent).
Analysis
Descriptive characteristics of the sample were examined through frequencies and means of the study variables. Bivariate associations between age norms and the study variables were investigated. Age group differences for age norms, level of prejudice against older adults and generational conflict were examined using ANOVA. Three regression models were developed to investigate age norms associated with prejudice against older adults and generational conflict among Koreans and whether these associations are conditioned by age. Specifically, a baseline model (Model 1) examines the association between age norms and demographic background variables including socio-economic status, social group participation and self-rated health status. A second model (Model 2) investigates the associations between age norm and both prejudice against older adults and generational conflict, controlling for the covariates. A final model (Model 3) adds two interaction terms to test moderation in the associations between age norm and both prejudice against older adults and generational conflictive perspective by age groups. Post hoc tests for overall group differences were then performed as separate analyses.
Results
Table 1 shows the characteristics of the sample by means and percentage distributions of the study variables as well as by their associations with age norms. Men and women were evenly distributed and approximately three-quarters of the sample were married. Almost 30 per cent of the respondents had a college-level education or more and about half of the sample were working. About 45 per cent of the sample did not have social meetings that they reported attending regularly, 20 per cent responded that they were attending one social meeting on a regular basis and the remainder indicated that they attended more than two social meetings regularly. The majority of respondents indicated that they were in good health, 20 per cent reported poor or fair health and the remainder reported being in excellent health. As presented in the last column of Table 1, almost all the potential covariates including age, education, employment status, household income, number of social meetings and self-rated health status showed significant associations with age norms at the bivariate level.
Table 1. Frequencies and means of the variables in the model and their associations with age norm
Notes: N = 1,445. SD: standard deviation.
Table 2 reports the levels of age norms, prejudice against older adults and generational conflictive perspective in the sample. The mean of the age norm scale was 3.10, indicating that the respondents tended to respond above ‘neutral’ in the scale ranges from 1 to 5. This result suggests that the respondents are inclined toward more restrictive perspectives on older adults behaviours overall. For prejudice against older adults and generational conflict, the full sample means were 2.33 and 2.87, respectively. When age groups were compared, older respondents demonstrated more restrictive attitudes about what is considered appropriate behaviour in later life, while prejudice against older adults and generational conflict appeared to be similar across the age groups.
Table 2. Age norm, prejudice toward older adults and generational conflict in the sample
Note: 1. Value for age group difference test.
Table 3 displays results from the regression models that examined the associations between age norms, prejudice against older adults and a generational conflictive perspective, as well as the other study variables. In Model 1, respondents 65 and older were more likely to support restrictive attitudes about acceptable behaviours in later life than those aged between 20 and 44 (the reference group), while those with better self-rated health showed less restrictive views on the behaviour of older adults. In Model 2, which included prejudice against older adults and generational conflictive perspective, respondents with greater prejudice tended to support more constraints on older adults’ behaviours (b = 0.57, β = 0.45, p = 0.000). The generational conflictive perspective, on the other hand, was not found to be associated with age norms at a 0.05 significance level. Age, social group meetings and self-rated health status continued to be significant determinants of age norms. Respondents who reported currently working at the time of interview showed less restrictive perceptions in determining what is the appropriate behaviour of older adults.
In Model 3, significant interactions between age and prejudice against older adults and generational conflictive perspective were found. Due to the nature of our grouping variable (i.e. age), which has three categories, two coefficients were estimated, each of which compares the association between age norms and prejudice against older adults or generational conflictive perspective with the reference group (i.e. age 20–44). The results indicate that the association between age norms for older adults and prejudice for the middle age group is significantly smaller than that for the younger adult group (i.e. the reference group; b = −0.15, β = −0.42, p = 0.034), while no group difference was found between the younger and the older adults groups. Another significant interaction between age and generational conflict was found: the association between generational conflictive perspective and age norms was smaller for the older adults group compared to the younger adults group (b = −0.17, β = −0.60, p = 0.008), while no difference was found between the younger and the middle-aged groups. Respondents who attended more social group meetings, were in better health and who were working consistently showed less restrictive attitudes in determining appropriate behaviours of older adults in this final model. Post hoc tests of overall group differences indicated that the associations between age norms and generational conflictive perspective were significantly different across the three age groups (F(2, 1426) = 8.46, p = 0.000), while the associations between age norm and prejudice against older adults were not. Based on the results of the post hoc analysis, we further performed three separate regression analyses on the association between age norms and generational conflict across the age groups. The results indicated that for middle-aged respondents, those who have a greater generational conflict perspective tend to place more constraints on older adults’ behaviours (b = 0.15, β = 0.15, p = 0.001), while those with more generational conflictive perspective supported less restriction in determining older adults’ appropriate behaviours among the older sample (b = −0.12, β = −0.10, p = 0.012). No association between age norm and generational conflict was found for the younger group of respondents.
Table 3. Regression models on age norm and demographic, socio-economic, social and health variables
Notes: N = 1,455. SE: standard error. Ref.: reference group.
Significance levels: * p < 0.5, ** p < 0.01, *** p < 0.001.
Discussion
The present study examined first, the degree of restrictions placed on age-appropriate behaviours for older adults among Koreans, secondly, influences of prejudice against older adults and generational conflictive perspectives in dealing with issues in an ageing society on these perceptions, and finally, potential age group differences in these relationships.
Study respondents tended to report a more restrictive perspective in determining what is the appropriate behaviour for older adults and being consistent with the existing literature (Lee Reference Lee2006), the older adults group appeared to give more support for this view. Previous research has noted that negative ideas about older adults are prevalent among older adults themselves (Giles and Reid Reference Giles and Reid2005; Levy Reference Levy2001) and this may occur as a result of lifetime exposure to ageism (Levy Reference Levy2001). We believe the reason for the greater age-structuring about what they should do or should not do among older Koreans themselves could be explained by Korean traditional culture that emphasises the family over individuals and this in turn may lead to more constraining attitudes concerning acceptable behaviour in later life. Older Koreans have been reported to place more value on social support, being respected by others and having harmonious family relationships as components of successful ageing rather than continued pursuit of individual wellbeing into old age (Paik Reference Paik2008; Paik and Choi Reference Paik and Choi2007). Given that studies on successful ageing in Western societies have reported that older adults tend to emphasise maintaining autonomy, control and vigour as factors defining successful ageing (Rowe and Kahn Reference Rowe and Kahn1997), studies that compare beliefs about age-appropriate behaviours for older adults across different cultures might help further elucidate how age norms are formed and how older adults perceive them.
A significant finding is that the association between a generational conflictive perspective and perceptions about age-appropriate behaviours of older adults was conditioned by the age of respondents. In the middle age group, those with a greater generational conflictive perspective were more likely to support restrictive attitudes in determining appropriate behaviours of older adults, while in the older adults group, a generational conflictive perspective was negatively associated with constraining attitudes on acceptable behaviours in later life. The non-significant association between generational conflictive perspective and age norms for older adults in the model without the interactions thus might be interpreted as the group differences in opposite directions which would have cancelled out the association of each other. We believe these age group differences can be understood in the context of the growing generational conflict in Korean society. As noted earlier, political discourse about the increasing social and economic cost of long-term care, generational equity in the operation of the national retirement pension system, and competition over employment between younger and older generations has become intensified (G. Park Reference Park2011; J. Park Reference Park2009). Middle-aged people in Korean society are the current contributors to the national retirement pension system and care-givers of their older parents.
On the one hand, given the increasing concern about their benefits compared to the contribution in the face of serious population ageing, those who tend to support policies benefiting younger generations may want to restrict the flexibility of behaviours and roles of older adults in society. On the other hand, the association between biased beliefs about older adults and increasingly restrictive ideas about appropriate behaviour in later life appeared similar across the age groups, as suggested by the non-significant post hoc statistical test of the overall group. This finding implies that prejudice against older adults may reinforce age-structuring in the perceived appropriateness of behaviours by older adults regardless of age. Although an overall group difference was not found, the significant difference in the association between age norms and prejudice against older adults between the younger and middle-aged respondents revealed in separate regressions should be noted. The biased negative beliefs about older adults held by younger adults exerted a stronger influence in their constraining perspectives of the acceptable behaviour of older adults. There has been a lack of multi-generational experience under the nuclear family structure in Korea and the finding may imply a need for an intergenerational programme designed to promote understanding between young and older generations (Choi et al. Reference Choi, Cheong, Cho and Jin2009; Choi and Han Reference Choi and Han2007; Ng Reference Ng1998).
Age-structured attitudes regarding the behaviour and roles of older adults may influence formal opportunity structures in employment, retirement and welfare programmes for older adults as these attitudes determine formally and informally expectations. For older adults themselves, restrictive attitudes about age-appropriate behaviours may become psychological impediments when seeking employment, extending social participation and engaging in self-development such as life-long learning.
The findings of our study reveal that negative views and expectations that older adults should withdraw from society, together with the presence of prejudice and generational conflict, contribute to determine acceptable behaviours of older adults in Korea. These findings suggest that the political effort to encourage productive ageing might be effective when older adults are specifically targeted. The findings give credence to the concept of an age-integrated society as a prerequisite of policies and programmes for older adults. Age integration is considered as a state in which no structural age barrier in work, education, family and leisure exists and entry to these institutions is flexible and open to all ages (Riley and Riley Reference Riley and Riley2000). In an age-integrated society, interactions across different ages are achieved by all ages (Riley and Riley Reference Riley and Riley2000; Uhlenberg Reference Uhlenberg2000). At an individual level, intergenerational interactions may contribute to a reduction of prejudice against older adults in younger generations that has often resulted from a lack of personal interaction with older adults (Won Reference Won2004). At a societal level, age integration would relax the institutionalisation of the lifecourse, which may allow age-normative and acceptable behaviours to be defined differently, thus enabling individuals to have a greater choice about their lifecourse in terms of the timing of education, work, retirement and leisure, based on their needs and preferences (Chung Reference Chung2004).
The limitations of the current study should be noted. First, the cross-sectional design of the study cannot capture the trajectories of age norms for older adults in changing Korean society. The present analysis only provides a one-shot picture of the association between constraint placed on older adults' behaviours and prejudice against older adults and a generational conflictive perspective despite the fact that these variables should be treated as dynamic. Future studies should aim to examine longitudinal relationships of these constructs which should provide more accurate empirical information for relevant policies and programmes.
Secondly, our measure of age norms does not encompass all of its aspects. As previously noted, age norms can be studied in terms of collective expectations about the occurrence of certain life transitions, and the degree of statistical regularity in age-related life events (Settersten Reference Settersten, Mortimer and Shanahan2003; Settersten and Mayer Reference Sherradan, Morrow-Howell, Hinterlong, Rozario, Morrow-Howell, Hinterlong and Sherradan1997). Our study does not examine whether there is a shared notion of a social clock for life transitions or how much consensus exists about such a life schedule. Our study instead focused on investigating the degree of restriction placed on perceptions of age-appropriate behaviours especially regarding social participation and engagement, and factors related with such perceptions among Koreans. Given that the concept of age norms in Korean society has remained under-researched, other aspects of age norms discussed in the literature should be examined.
Acknowledgement
This work was supported by the National Research Foundation of Korea Grant funded by the Korean Government (NRF-2010-330-B00209).
Appendix
Survey questions on age norm among Koreans
