Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Regional Impact of Population Aging on Changes in Individual Self-perceptions of Aging: Findings From the German Ageing Survey

Regional Impact of Population Aging on Changes in Individual Self-perceptions of Aging: Findings... Abstract Background and Objectives The importance of self-perceptions of aging (SPA) for health and longevity is well documented. Comparably little is known about factors that contribute to SPA. Besides individual factors, the context a person lives in may shape SPA. Research has so far focused on country-level differences in age stereotypes, indicating that rapid population aging accompanies more negative age stereotypes. The present study expands previous research by investigating the impact of district-specific population aging within one country on different facets of SPA. Research Design and Methods Based on a large representative survey in Germany, the study investigates changes in SPA as ongoing development as well as the SPA of physical loss over a 12-year period in adults aged 40+. The study uses several indicators of population aging (e.g., population development, average age, greying index), to identify four clusters differing in their pace of population aging. Based on three-level latent change models, these clusters were compared in their impact on changes in SPA. Results Compared to districts with an average rate of population aging, the study shows that persons living in regions with a fast population aging rate (C1) hold more negative SPA in both facets (ps = .01). Districts with slow population aging (C2) have significantly higher SPA ongoing development (p = .03). Discussion and Implications The study underlines the importance for regional differences in population aging on the development of SPA. In particular, societies should be aware that fast population aging may result in more negative SPA. Attitudes and perception toward aging/aged, Demography, context factors/contextual influences, Successful aging A growing body of research underlines the importance of investigating contextual influences on psychological phenomena such as attitudes and values (Rentfrow & Jokela, 2016). Gerstorf and Ram (2012) collected empirical evidence on the effects of contextual factors on developmental trajectories in later life. The authors state that four components of context may be linked to individual development: economic characteristics (e.g., level of poverty or rate of unemployment), service characteristics (e.g., availability and accessibility of health services), social characteristics (e.g., social ties or perceived discrimination), and physical characteristics (e.g., features of the built environment or accessibility). Aging well is thus shaped by individual resources and constraints as well as contextual factors on various levels (e.g., country, region, and neighborhood). An important indicator of aging well is the individual view on one’s own aging process. Numerous studies have illustrated the far-reaching importance of self-perceptions of aging (SPA) for health, well-being, and longevity (Westerhof et al., 2014). However, little is known about which factors, besides individual characteristics, may shape SPA. Considering the speed of current demographic change and population aging in most industrialized countries, their impact on SPA should clearly be assessed. Thus, the current study investigates the effect differences in the pace of population aging on SPA have regionally, over 12 years. Determinants of SPA In accordance with stereotype embodiment theory (Levy, 2009) and supported by empirical findings (e.g., Kornadt, Voss, & Rothermund, 2017), age stereotypes are learned and internalized during youth, accumulate over the life span, and become SPA when they are directed at oneself and the own aging process with increasing age. Changes in SPA can be investigated from two perspectives: intraindividual and social change. A recent study on social change in SPA in the German population between 1996 and 2014 showed that SPA have improved in persons aged 40–85 years (Beyer, Wurm, & Wolff, 2017). Longitudinal studies on intraindividual changes in SPA with age are rare (e.g., Kleinspehn-Ammerlahn, Kotter-Grühn, & Smith, 2008) but show, in line with cross-sectional age group comparisons, that older individuals report more negative SPA than younger individuals (e.g., Wurm, Wolff, & Schüz, 2014). Intraindividual and social change of age stereotypes and SPA can be influenced by individual factors such as socioeconomic variables (e.g., Kleinspehn-Ammerlahn et al., 2008), health status (e.g., Sargent-Cox, Anstey, & Luszcz, 2012; Wurm, Tesch-Römer, & Tomasik, 2007), personality (e.g., Bryant et al., 2016; Levy, 2008; Moor, Zimprich, Schmitt, & Kliegel, 2006), and own and vicarious aging experiences, such as receipt of care (e.g., Kwak, Ingersoll-Dayton, & Burgard, 2014). Contextual influences may be defined as “(…) the totality of the diverse range of phenomena, events, and forces that exist outside the developing individual, including its sociocultural and physical aspects” (Dannefer, 1992, p. 84). Additionally, the environment a person lives in changes over time (Bronfenbrenner, 1977), and, according to the ecological systems theory (Bronfenbrenner, 1979) and the cross-level dynamic biocultural coconstructive framework of development (Li, 2003), contextual variables influence behavior and attitudes of individuals. Empirically, contextual variables such as the portrayal of age and older adults in the media (e.g., Donlon, Ashman, & Levy, 2005), age-graded social regulations such as retirement (e.g., Kim & Moen, 2002), or the societal roles of older adults (e.g., Bowen & Skirbekk, 2013) have been shown to play a role in shaping age stereotypes and their internalization into SPA. However, even though many western societies face demographic changes, characteristics of population aging at the regional level have been rarely examined as contextual influences on SPA. Population Aging and SPA While the effect of population aging on SPA has not been examined so far, first evidence suggests that population aging affects age stereotypes and attitudes towards aging. A comparison of eastern and western cultures showed overall more favorable age stereotypes in western cultures (e.g., Giles et al., 2003; McCann & Keaton, 2013), which seems to be accounted for by differences in population aging (see meta-analysis by North & Fiske, 2015 for details). North and Fiske (2015) showed that individuals living in countries with rapid population aging, which is true for most eastern countries, have more negative age stereotypes. In a similar vein, Löckenhoff et al. (2009) compared aging perceptions of mostly younger adults in 26 countries and found that countries with a higher proportion of inhabitants aged 65+ had more negative perceptions of aging. These findings, however, contradict the assumption of the contact hypothesis (cf. Allport, 1954; Pettigrew & Tropp, 2006), which suggests that having more contact with a group of people will result in reduced prejudices towards that group. That is, individuals living in contexts with comparatively more older adults should have fewer prejudices toward older adults, which could, in turn, be associated with decreased negative age stereotypes. However, these studies focused on age stereotypes or attitudes toward aging and did not investigate how an individual perceives his or her own aging process. Regarding the relation between population aging and SPA, there are good arguments for both more positive and more negative SPA due to population aging: On the one hand, increased contact with older adults due to living in an area with a larger older adult population could reduce negative SPA as population aging may enhance the likelihood of experiencing healthy older adults, who actively participate in society. On the other hand, population aging can impose threats and challenges on a society. For example, anticipated problems in distributive justice of health care and pension systems, increased care needs of old-old individuals, or the financial burden of younger family members or the younger generation in general may result in more negative age stereotypes and SPA. In particular, rapid population aging may hamper adaptation to these changes. Silverstein, Parrott, Angelelli, and Cook (2000) argue that a greater proportion of older adults may result in intergenerational conflict over limited resources. Cagny and Wen (2008) state that age structure drives the demand for services (e.g., health clinics) and infrastructure (e.g., sidewalks) as well as fostering expectations about social roles (e.g., informal monitoring of children) that may affect social contact and cohesion in neighborhoods. Putnam (2000) found that a high number of aged residents is associated with a more active neighborhood watch, better social services, and, in general, a community more engaged in civic affairs. Furthermore, social characteristics on a regional level affect individuals via informal social control (e.g., monitoring, expectations for action, availability of role models; cf. Cagny & Wen, 2008; Gerstorf & Ram, 2012). These changes in infrastructure and norms may provide more opportunities for intergenerational contact resulting in stronger social cohesion. For example, increased intergenerational contact due to a community program was already shown to result in more positive images of aging in fourth grade students (even though no effect was found for negative images of aging; Thompson & Weaver, 2016). A country level perspective, however, ignores heterogeneity in population aging within a country. Aspects of population aging, such as increases in the proportion of elderly population, do not happen in the same manner in all geographical regions of one country (Menning, Nowossadeck, & Maretzke, 2010). In addition, regional patterns of population aging affect more directly the immediate environment where people live as well as the perception of older people and may, therefore, have stronger effects on SPA. In Germany, districts are responsible for various infrastructural aspects, such as establishing economic infrastructures, planning for housing areas, providing health care services, and offering opportunities for volunteering. Previous research has largely neglected regional differences in population aging, age stereotypes, and SPA, even though there are some hints of regional differences in age stereotypes for example between urban and rural areas (favoring rural areas; Macia, Lahmam, Baali, Boëtsch, & Chapuis-Lucciani, 2009). Thus, the analysis of population aging at the district-level provides additional information to previous studies on country-level aging: First, the heterogeneity of population aging within one country is captured, and, second, this perspective takes into account that age stereotypes and SPA emerge under specific regional circumstances of social, economic, and cultural life. The Present Study No study has thus far examined the impact of population aging on SPA. The present study therefore explores the question whether district level differences in population aging in Germany explain changes in SPA. Population aging is an interaction of different demographic factors and a complex phenomenon that cannot be captured sufficiently with single indicators, such as percentage of older adults (as used in most of the previous studies). Therefore, we decided to use four clusters reflecting differences in indicators of population aging (e.g., average age; proportion of very old individuals [80+] at district level) were used in the analyses (see Method section for details; Menning et al., 2010). The clusters should explain variability in change in two different facets of SPA: the experience of aging as both ongoing development and physical loss. These two facets capture two important dimensions of the multidimensional construct of SPA (cf. Kornadt & Rothermund, 2011; Steverink, Westerhof, Bode, & Dittmann-Kohli, 2001; Wurm et al., 2007). Ongoing development describes the belief that aging is accompanied with new opportunities and one is still making new plans. Chances for ongoing development are affected by contextual factors such as opportunities for leisure or other activities. The second facet captures the common belief that aging is strongly associated with physical loss. This loss-related view might also be prone to contextual influences, for example, in terms of environments equipped for older and functional disabled persons. This study examines population aging at the district level and changes in two facets of SPA across 12 years using data from a large representative survey of adults aged 40 years and older in Germany to answer the research question: Do different regional patterns of population aging influence changes in SPA? Design and Methods Sample Data come from two measurement points (1996, 2008) of the German Ageing Survey (DEutscher AltersSurvey [DEAS]; Klaus et al., 2017). DEAS is a national representative cohort-sequential study of adults aged 40 years and older. Cross-sectional baseline samples have been drawn every 6 years since 1996 and are stratified by age, gender, and region (Eastern/Western Germany). The sampling procedure of the DEAS starts with randomly drawn municipalities, in which stratified samples are randomly drawn from registry offices. Thereby, the sample is representative of the German population, with only a subsample of all 413 German districts in the data. At each measurement point, participants took part in a 90-min interview and filled in a paper-pencil questionnaire. The current study combines cross-sectional and longitudinal information from two measurement points (t1: 1996 and t2: 2008; see Figure 1) to assess a combination of social change (between baseline samples of 1996 and 2008) and intraindividual change (between t1 [1996] and t2 [2008]) in SPA. The 1996 sample data represents a baseline at t1 (n = 4,838). The 2008 sample includes both baseline data from additional participants (n = 6,205) as well as the longitudinal data from the 1996 participants. Due to political reforms that led to changes in the number of districts in Germany, 155 persons had to be excluded because they could not be clearly assigned to one of the clusters (n1996 = 32; n2008 = 123). As SPA were assessed in the paper-pencil questionnaire, only participants who completed the questionnaire were included in the analyses: Of the remaining sample from 1996, n = 797 persons did not fill-in the paper-pencil questionnaire in 1996 and n = 148 in 2008. Of the remaining sample in 2008, n = 1,699 participants had to be excluded due to missing paper-pencil questionnaires. Additionally, we excluded all persons who moved between districts between measurement points (n = 45 of the 1996 sample). The analyzed sample in 1996 includes 3,816 persons. The analyzed sample in 2008 consists of 5,122 persons, including information from the 2008 baseline sample (n = 4,383) and longitudinal information from the 1996 sample (n = 739). Thus, a total sample of n = 8,199 was analyzed. Of the 413 districts in Germany, 207 are represented in the data. On average, 40 persons lived in each district. Figure 1. View largeDownload slide Flow-chart of DEAS participants used in the study; p&p = paper & pencil questionnaire; B1996 = baseline sample 1996; L2008 = longitudinal sample of baseline participants 1996; B2008 = baseline sample 2008. DEAS = DEutscher AltersSurvey. Figure 1. View largeDownload slide Flow-chart of DEAS participants used in the study; p&p = paper & pencil questionnaire; B1996 = baseline sample 1996; L2008 = longitudinal sample of baseline participants 1996; B2008 = baseline sample 2008. DEAS = DEutscher AltersSurvey. Measures Demographic Clusters Four clusters differing in their pace of population aging were identified by a cluster analysis (Menning et al., 2010) aimed at describing population aging in all German districts based on the following indicators: (a) population development per 1,000 inhabitants between 1995 and 2008, (b) average age of population in 2008, (c) percentage of population aged 65+ in 1995, (d) percentage of population aged 65+ in 2008, (e) greying index (old-old population, aged 80+ years, in relation to young-old, aged 65–79 years) in 2008, and (f) balance of births and deaths per 1,000 inhabitants in 2008. The average values of these indicators from each cluster for all German districts are summarized in Table 1 and described below. Table 1. Cluster Indicators of Population Aging in Germany (average values, 413 districts in 2008) Indicators  Cluster 1 fast aging; 2008: highest average age/low proportion of 80+  Cluster 2 slow aging; 2008: lowest average age/low proportion of 80+  Cluster 3 average aging; 2008: high average age/high proportion of 80+  Cluster 4 average aging; 2008: average age/average proportion of 80+  Population development per 1,000 inhabitants, between 1995 and 2008 (balance)  −119.7  +70.5  −31.3  +17.1  Average age of population in 2008 (years)  45.8  42.0  44.3  43.0  Share of population aged 65 years and older in 2008 (%)  23.5  18.6  22.4  20.3  Share of population aged 65 years and older in 1995 (%)  14.9  13.8  18.4  15.9  Greying index (very old population aged 80 years and older in relation to the young elderly aged from 65 to 79 years) in 2008 (ratio)  27.6  30.3  37.1  33.4  Balance of births and deaths per 1,000 inhabitants in 2008 (birth surplus)  −4.5  −0.4  −4.4  −2.2  Indicators  Cluster 1 fast aging; 2008: highest average age/low proportion of 80+  Cluster 2 slow aging; 2008: lowest average age/low proportion of 80+  Cluster 3 average aging; 2008: high average age/high proportion of 80+  Cluster 4 average aging; 2008: average age/average proportion of 80+  Population development per 1,000 inhabitants, between 1995 and 2008 (balance)  −119.7  +70.5  −31.3  +17.1  Average age of population in 2008 (years)  45.8  42.0  44.3  43.0  Share of population aged 65 years and older in 2008 (%)  23.5  18.6  22.4  20.3  Share of population aged 65 years and older in 1995 (%)  14.9  13.8  18.4  15.9  Greying index (very old population aged 80 years and older in relation to the young elderly aged from 65 to 79 years) in 2008 (ratio)  27.6  30.3  37.1  33.4  Balance of births and deaths per 1,000 inhabitants in 2008 (birth surplus)  −4.5  −0.4  −4.4  −2.2  View Large Cluster 1 (C1; fast aging; 2008: highest average age/low proportion of 80+) Cluster 1 comprises districts with a rapidly aging, shrinking population. All districts in this cluster are located in Eastern Germany. In the past, these districts had a relatively young population structure. Since 1995 they have been confronted with population decline and an accelerated aging process. The average age in 2008 is the highest of all clusters. The percentage of people aged 65+ is above average, although the percentage of people aged 80+ is the smallest among all clusters. In this study, 42 districts with 2,084 participants belong to C1. Cluster 2 (C2; slow aging; 2008: lowest average age/low proportion of 80+) The districts of this cluster are mostly situated in urban agglomerations and are characterized by a slowly aging, growing population. The population has been increasing since 1995, mostly due to internal migration. The average age in 2008 is lower than the average for Germany. Moreover, the percentages of people aged 65+ or 80+ are low in comparison to the other clusters. The districts of C2 have stable and growing populations. Demographic change is occurring at a much slower pace. In this study, 53 districts with 2,266 participants belong to C2. Cluster 3 (C3; average aging; 2008: high average age/high proportion of 80+) This cluster comprises districts with a slowly aging, shrinking population. All districts in this cluster are located in Western Germany. Population aging in this cluster has progressed considerably: In contrast to C1 it started with a much older population in 1995; since then, the pace of population aging has been slower than in C1. The percentage of the population aged 65+ is higher than on average. The greying index is the highest of all clusters in 2008. In this study, 31 districts and 849 participants belong to C3. Cluster 4 (C4; average aging; 2008: average age/average proportion of 80+) Cluster 4 consists of districts with an average rate of population aging. The vast majority of districts in this cluster are situated in Western Germany. They are characterized by a medium pace of population aging and minimal population growth since 1995. Most of the indicators of population aging are close to the average for Germany. In this study, 81 districts and 3,000 participants belong to C4. For the analyses, the clusters were dummy coded using C4 as reference category as it represents average population aging in Germany. SPA Two scales of SPA were assessed in 1996 and 2008 with subscales for the aging-related cognition of ongoing development and physical loss (AgeCog-Scales; Steverink et al., 2001; Wurm et al., 2007). The SPA ongoing development consists of four items and refers to the positive view that aging is seen as a time of continuous personal development (e.g., “Aging means to me that I can still learn new things,” “Aging means to me that my capabilities are increasing”; Cronbach’s α1996 = .91; Cronbach’s α2008 = .93). The SPA physical losses consists of four items and addresses the view that aging is accompanied by physical decline (e.g., “Aging means to me that I am less healthy,” “Aging means to me that I am less energetic and fit”; Cronbach’s α1996 = .91; Cronbach’s α2008 = .91). Responses could range from 1 (strongly disagree) to 4 (strongly agree). For each scale, the four items were averaged to obtain a score, with higher values indicating greater agreement. That is, a higher SPA score for ongoing development indicates a more positive view, and a higher SPA score for physical losses indicates a more negative view. Covariates (individual) Covariates on the individual level were age, gender, and region (Eastern/Western Germany) as the DEAS sample is stratified by theses variables and to account for age-related differences in SPA. As more healthy and highly educated people are known to report more positive SPA, analyses were also controlled for education. This was assessed using the International Standard Classification of Education (ISCED with three levels; UNESCO, 2012). The three levels are: low education (9 years of school education at most), medium education (secondary school), and high education (qualifying for university admission). Additionally, the health status of the participants was controlled for using both self-reported numbers of physical diseases (up to 11 diseases such as cardiovascular diseases, back or joint diseases, respiratory diseases) and functional health (physical functioning subscale of the SF-36 questionaire; Bullinger & Kirchberger, 1998). The latter was coded according to the manual, with values ranging from 1 to 100 and higher values indicating better functional health. Values of these covariates are based on the first assessment point of each participant. Covariates (district) The variables included as covariates are indicators of the wealth and health care opportunities of the district, which have been shown to be related to SPA and psychological resources (Wurm et al., 2014). On a district level, all analyses were controlled for using the percentage of unemployed older workers (55–65 years) from all unemployed persons (M [SD] = 13.56 [2.31]), primary care supply (number of general practitioners per 100,000 inhabitants; M [SD] = 673.30 [179.51]), gross domestic product (GDP in 1000 Euro per capita; M [SD] = 28.66 [12.37]), and population density (number of residents per km2; M [SD] = 749.82 [967.97]) in 2008. In addition, the analyses were controlled for using the district level average remaining life expectancy for 60-year-old persons between 2007 and 2009 (M [SD] = 23.21 [0.74]). Primary care supply, GDP, and population density were divided by 100 before including them in the analyses to facilitate the estimation of the models. Data Analyses The data has a three-level structure with time on Level 1, individuals on Level 2, and districts on Level 3. Therefore, a three-level latent change model was used to analyze the data in Mplus (adapted from Mun, von Eye, & White, 2009). Within-person change in SPA across 12 years was modeled, allowing for individual and district-level differences in this change as well as in intercept. Change in SPA ongoing development and SPA physical losses were estimated in two separate models. The intercept was modeled to represent levels at 2008 allowing for testing level differences in SPA in 2008. Intercepts and change scores were predicted by the cluster dummies (C4 as reference) and all covariates. Estimation was conducted with full information maximum likelihood estimation (FIML) to account for longitudinal drop-out in the sample (cf. Schafer & Graham, 2002). Significance level was set to .05. All Level 2 variables were group-mean centered. All Level 3 variables were grand-mean centered. Covariates were allowed to covary on individual and district level. To ensure model identification, the residual variances of SPA ongoing development and SPA physical losses were set zero for 1996, and the intercept and change scores were not allowed to covary on individual level. Results Descriptive Statistics The average age at first measurement point was 61.06 years (SD = 12.05, range: 40–85 years) and 48.70% were female. Of all participants, 12.96% had low, 56.04% medium, and 31.00% high-level education. On average, the score on the physical functioning subscale of the SF-36 was M (SD) = 83.99 (22.37) and participants reported 2.41 diseases (SD = 1.87). SPA Ongoing Development The intraclass correlation coefficient (ICC: the proportion of variance on district level to total variance) for SPA ongoing development was .05 and .06 in 1996 and 2008, respectively. The results of the multilevel model are summarized in Table 2. Table 2. Unstandardized Parameters of Multilevel Latent Change Model for SPA Ongoing Development using DEAS Data From 1996 to 2008 Parameter  Estimate (SE)  p value  Fixed Effects (Level 1 Time)   Intercept (level 2008)  2.95 (0.03)  <.001   Change (1996–2008)  −0.03 (0.03)  .30  Fixed Effects (Level 2 Individual)  Intercept on       Age  −0.01 (0.001)  <.001   Gender  0.05 (0.02)  .001   Eastern/Western Germany  −0.12 (0.06)  .03   Education  0.13 (0.01)  <.001   Functional Health  0.01 (0.001)  <.001   Number of Diseases  −0.03 (0.01)  <.001  Change on       Age  −0.01 (0.001)  <.001   Gender  0.02 (0.02)  .42   Eastern/Western Germany  0.03 (0.05)  .51   Education  0.003 (0.02)  .89   Functional Health  −0.003 (0.001)  <.001   Number of Diseases  0.01 (0.01)  .38  Fixed Effects (Level 3 District)  Intercept on       C1 vs C4  −0.10 (0.04)  .01   C2 vs C4  0.03 (0.04)  .50   C3 vs C4  0.01 (0.04)  .88   Unemployment Rate 2008  −0.001 (0.01)  .89   Primary Care Supply 2008  −0.01 (0.01)  .13   GDP 2008  −0.08 (0.13)  .54   Remaining Life Expectancy 2008  0.01 (0.02)  .59   Population Density 2008  −0.002 (0.003)  .56  Change on       C1 vs C4  −0.04 (0.05)  .38   C2 vs C4  −0.10 (0.05)  .03   C3 vs C4  −0.04 (0.05)  .46   Unemployment Rate 2008  −0.002 (0.01)  .75   Primary Care Supply 2008  0.01 (0.01)  .75   GDP 2008  −0.01 (0.15)  .94   Remaining Life Expectancy 2008  −0.02 (0.03)  .49   Population Density 2008  0.002 (0.003)  .55  Random Effects (Level 2 Individual)   Intercept  0.13 (0.02)  <.001   Change  0.24 (0.02)  <.001  Random Effects (Level 3 District)   Intercept  0.02 (0.01)  <.001   Change  0.01 (0.01)  .11  Random Effects (Level 1 Time)   Residual  0.17 (0.02)  <.001  Parameter  Estimate (SE)  p value  Fixed Effects (Level 1 Time)   Intercept (level 2008)  2.95 (0.03)  <.001   Change (1996–2008)  −0.03 (0.03)  .30  Fixed Effects (Level 2 Individual)  Intercept on       Age  −0.01 (0.001)  <.001   Gender  0.05 (0.02)  .001   Eastern/Western Germany  −0.12 (0.06)  .03   Education  0.13 (0.01)  <.001   Functional Health  0.01 (0.001)  <.001   Number of Diseases  −0.03 (0.01)  <.001  Change on       Age  −0.01 (0.001)  <.001   Gender  0.02 (0.02)  .42   Eastern/Western Germany  0.03 (0.05)  .51   Education  0.003 (0.02)  .89   Functional Health  −0.003 (0.001)  <.001   Number of Diseases  0.01 (0.01)  .38  Fixed Effects (Level 3 District)  Intercept on       C1 vs C4  −0.10 (0.04)  .01   C2 vs C4  0.03 (0.04)  .50   C3 vs C4  0.01 (0.04)  .88   Unemployment Rate 2008  −0.001 (0.01)  .89   Primary Care Supply 2008  −0.01 (0.01)  .13   GDP 2008  −0.08 (0.13)  .54   Remaining Life Expectancy 2008  0.01 (0.02)  .59   Population Density 2008  −0.002 (0.003)  .56  Change on       C1 vs C4  −0.04 (0.05)  .38   C2 vs C4  −0.10 (0.05)  .03   C3 vs C4  −0.04 (0.05)  .46   Unemployment Rate 2008  −0.002 (0.01)  .75   Primary Care Supply 2008  0.01 (0.01)  .75   GDP 2008  −0.01 (0.15)  .94   Remaining Life Expectancy 2008  −0.02 (0.03)  .49   Population Density 2008  0.002 (0.003)  .55  Random Effects (Level 2 Individual)   Intercept  0.13 (0.02)  <.001   Change  0.24 (0.02)  <.001  Random Effects (Level 3 District)   Intercept  0.02 (0.01)  <.001   Change  0.01 (0.01)  .11  Random Effects (Level 1 Time)   Residual  0.17 (0.02)  <.001  Note: DEAS = DEutscher AltersSurvey; GDP = Gross domestic product; SPA = Self-perceptions of aging. View Large In 2008, the average value of SPA ongoing development was 2.95 in the reference group (C4). Individuals living in C1 reported a lower level of SPA ongoing development than C4 (bC1 = −0.10, p = .01). C2 as well as C3 had a similar level compared to C4 because there was no significant difference in the intercept (bC2 = 0.03, p = .50; bC3 = 0.01, p = .88). The cluster variables explained 4.55% of the district-level variance in the intercept. Regarding changes from 1996 to 2008 (see Figure 2), the average value of SPA ongoing development remained stable over time in the reference group C4 (change = −0.03, p = .30). C1 and C3 did not differ from C4 in changes of SPA ongoing development (bC1 = −0.04, p = .38; bC3 = −0.04, p = .46), meaning SPA also remained stable over time in C1 and C3. Change over time differed between C2 and the reference group C4, as indicated by the significant effects of C2 on the change score (bC2 = −0.10, p = .03). That is, individuals living in C2 reported more positive SPA in 2008 than in 1996. The cluster variables explained 9.09% of the district-level slope variance. Figure 2. View largeDownload slide Change in SPA ongoing development from 1996 to 2008 in clusters of population aging using DEAS data. DEAS = DEutscher AltersSurvey; SPA = Self-perceptions of aging. Figure 2. View largeDownload slide Change in SPA ongoing development from 1996 to 2008 in clusters of population aging using DEAS data. DEAS = DEutscher AltersSurvey; SPA = Self-perceptions of aging. In sum, C1 showed the lowest levels of SPA ongoing development, meaning more negative SPA were reported in C1 as compared to the other clusters in 2008. Only C2 showed a significant increase in SPA ongoing development over time. The other clusters remained stable. SPA Physical Losses Results of the multilevel model for SPA physical losses are shown in Table 3. The ICC was .02 for SPA physical losses in 1996 and .04 in 2008. Table 3. Unstandardized Parameters of Multilevel Latent Change Model for SPA Physical Losses using DEAS Data From 1996 to 2008 Parameter  Estimate (SE)  p value  Fixed Effects (Level 1 Time)   Intercept (level 2008)  2.65 (0.03)  <.001   Change (1996–2008)  −0.004 (0.03)  .90  Fixed Effects (Level 2 Individual)  Intercept on       Age  0.001 (0.001)  .11   Gender  −0.04 (0.02)  .02   Eastern/Western Germany  0.11 (0.07)  .10   Education  −0.04 (0.01)  .01   Functional Health  −0.01 (0.000)  <.001   Number of Diseases  0.06 (0.01)  <.001  Change on       Age  0.01 (0.001)  <.001   Gender  0.03 (0.03)  .34   Eastern/Western Germany  −0.12 (0.06)  .06   Education  −0.02 (0.02)  .47   Functional Health  0.004 (0.001)  <.001   Number of Diseases  0.03 (0.01)  <.001  Fixed Effects (Level 3 District)  Intercept on       C1 vs C4  0.09 (0.04)  .01   C2 vs C4  0.004 (0.03)  .89   C3 vs C4  0.02 (0.04)  .58   Unemployment Rate 2008  −0.01 (0.01)  .18   Primary Care Supply 2008  −0.002 (0.01)  .82   GDP 2008  −0.01 (0.11)  .91   Remaining Life Expectancy 2008  0.01 (0.02)  .56   Population Density 2008  −0.003 (0.003)  .25  Change on       C1 vs C4  −0.04 (0.05)  .44   C2 vs C4  0.05 (0.04)  .22   C3 vs C4  −0.06 (0.06)  .28   Unemployment Rate 2008  −0.01 (0.01)  .53   Primary Care Supply 2008  0.001 (0.01)  .96   GDP 2008  0.04 (0.16)  .82   Remaining Life Expectancy 2008  −0.03 (0.03)  .31   Population Density 2008  0.000 (0.003)  .99  Random Effects (Level 2 Individual)   Intercept  0.11 (0.01)  <.001   Change  0.23 (0.01)  <.001  Random Effects (Level 3 District)   Intercept  0.01 (0.003)  <.001   Change  0.02 (0.01)  .004  Random Effects (Level 1 Time)   Residual  0.14 (0.01)  <.001  Parameter  Estimate (SE)  p value  Fixed Effects (Level 1 Time)   Intercept (level 2008)  2.65 (0.03)  <.001   Change (1996–2008)  −0.004 (0.03)  .90  Fixed Effects (Level 2 Individual)  Intercept on       Age  0.001 (0.001)  .11   Gender  −0.04 (0.02)  .02   Eastern/Western Germany  0.11 (0.07)  .10   Education  −0.04 (0.01)  .01   Functional Health  −0.01 (0.000)  <.001   Number of Diseases  0.06 (0.01)  <.001  Change on       Age  0.01 (0.001)  <.001   Gender  0.03 (0.03)  .34   Eastern/Western Germany  −0.12 (0.06)  .06   Education  −0.02 (0.02)  .47   Functional Health  0.004 (0.001)  <.001   Number of Diseases  0.03 (0.01)  <.001  Fixed Effects (Level 3 District)  Intercept on       C1 vs C4  0.09 (0.04)  .01   C2 vs C4  0.004 (0.03)  .89   C3 vs C4  0.02 (0.04)  .58   Unemployment Rate 2008  −0.01 (0.01)  .18   Primary Care Supply 2008  −0.002 (0.01)  .82   GDP 2008  −0.01 (0.11)  .91   Remaining Life Expectancy 2008  0.01 (0.02)  .56   Population Density 2008  −0.003 (0.003)  .25  Change on       C1 vs C4  −0.04 (0.05)  .44   C2 vs C4  0.05 (0.04)  .22   C3 vs C4  −0.06 (0.06)  .28   Unemployment Rate 2008  −0.01 (0.01)  .53   Primary Care Supply 2008  0.001 (0.01)  .96   GDP 2008  0.04 (0.16)  .82   Remaining Life Expectancy 2008  −0.03 (0.03)  .31   Population Density 2008  0.000 (0.003)  .99  Random Effects (Level 2 Individual)   Intercept  0.11 (0.01)  <.001   Change  0.23 (0.01)  <.001  Random Effects (Level 3 District)   Intercept  0.01 (0.003)  <.001   Change  0.02 (0.01)  .004  Random Effects (Level 1 Time)   Residual  0.14 (0.01)  <.001  Note: DEAS = DEutscher AltersSurvey; GDP = Gross domestic product; SPA = Self-perceptions of aging. View Large In 2008, the average value of SPA physical loss was 2.65 in the reference group (C4). C2 and C3 did not differ in intercept from C4 (bC2 = 0.004, p = .89; bC3 = 0.02, p = .58), which means that individuals living in C2 and C3 reported similar levels for SPA physical losses, compared to individuals living in C4. However, C1 had higher levels of SPA physical losses in 2008 as compared to C4, as indicated by a significant effect of C1 on the intercept (b = 0.09, p = .01). In total, 7.69% of the intercept variance was explained by the cluster variables. Regarding changes from 1996 to 2008 (see Figure 3), the average value of SPA physical losses remained stable over time in the reference group (change = −0.004, p = .90). Figure 3. View largeDownload slide Change in SPA physical losses from 1996 to 2008 in clusters of population aging using DEAS data. DEAS = DEutscher AltersSurvey; SPA = Self-perceptions of aging. Figure 3. View largeDownload slide Change in SPA physical losses from 1996 to 2008 in clusters of population aging using DEAS data. DEAS = DEutscher AltersSurvey; SPA = Self-perceptions of aging. None of the other clusters differed in change over time from C4 (bC1 = −0.04, p = .44; bC2 = 0.05, p = .22; bC3 = −0.06, p = .28), indicating no significant change in SPA physical losses between 1996 and 2008 in C1, C2, or C3. In sum, C1 showed higher levels of SPA physical losses, which indicates more negative SPA in C1 as compared to the other clusters. SPA physical losses did not change over time in all clusters. Discussion This study is one of the first to investigate regional differences in SPA development by referring to population aging on a district level. Two different SPA were examined: Associating the own aging process with either ongoing development or with physical losses. Results show that population aging influenced the level of and changes in the SPA ongoing development as well as level in the SPA physical losses, whereas population aging did not predict change in the SPA physical losses. Consistently, individuals living in districts characterized by a fast aging population, high average age, and low proportion of 80+ (C1) reported more negative SPA for both facets in 2008. Change in SPA physical losses was not associated with differences in population aging, but individuals living in districts with slow population aging, low average age, and low proportion of 80+ (C2) showed a more positive development of SPA ongoing development over time as compared to individuals living in regions with average population aging (C4). In line with the findings of North and Fiske (2015), results suggest that persons have more negative SPA if they live in regions where population aging happens faster. A relatively fast change in the composition of a population may affect the perception of how well needs are met in regard to local circumstances. Changes in leisure time facilities, or regarding health care (e.g., because of retiring General Practitioners) and also the mere age-related changes in the social structure of the neighborhood may affect the living conditions of older adults in these areas and pronounce negative aspects and barriers in old age, resulting in more negative perceptions of aging. Investing in infrastructure and in leisure time facilities for older persons may contribute to the perception of having opportunities for ongoing development in old age among the inhabitants of these regions. However, to accomplish this, regions need to have sufficient financial and economic strength. Fast population aging in combination with low economic strength may even amplify negative changes in SPA. On an individual level, it has already been shown that financial resources make a difference for SPA and their consequences (Craciun, Gellert, & Flick, 2017). Vitman, Iecovich, & Alfasi (2014) also demonstrated that social integration of older adults is associated with characteristics of neighborhoods, such as socioeconomic status, ageism, or perceived limitations of outdoor mobility, in Tel-Aviv. There is, however, a need for more studies investigating the interplay of contextual variables and psychological phenomena. Also, positive effects of regional characteristics on SPA were observed in this study. Persons living in regions with slow population aging (e.g., relatively low average age and small proportion of old-old adults) as compared to the German average showed a positive development in their SPA ongoing development. Possibly, these regions are able to better adapt to a slowly growing number of older adults as compared to regions with faster population aging. A lower proportion of older adults, and in particular of old-old adults, may result in fewer negative role models and, therefore, more positive SPA. This would be in line with research showing, that perception of older adults depends on societal roles of older adults: Older persons are perceived as more competent in countries with a higher proportion of older adults participating in paid and volunteer work (Bowen & Skirbekk, 2013). In particular, regions with higher proportions of healthy older adults may benefit from these positive effects. Additionally, social contact and cohesion may be different in those communities where the population structure has been shifted gradually over time. As suggested in the model by Li (2003), intergenerational interactions may shape development of individuals. Future research should investigate which mechanisms might underlie this relationship. Change in SPA physical losses, in contrast, was not affected by population aging as operationalized in the clusters. Löckenhoff and colleagues (2009) argue that context effects are strongest for the aspects of aging perceptions that are least influenced by age-related biological changes, which was supported by the finding that physical perceptions of aging were less related to country level indicators than other facets such as socioemotional or family-related views of aging. Possibly, expectations of physical loss with aging are more likely to be influenced by individual factors such as health behaviors or health events than by contextual circumstances. In sum, our findings add to the previous research by extending what has been shown for age stereotypes in different countries (cf. North & Fiske, 2015): Population aging is associated with changes in self-perceptions of aging. To some extent, results contradict the prediction of the contact hypothesis (cf. Allport, 1954; Pettigrew & Tropp, 2006). However, little is known about mechanisms of population aging on social contact and cohesion. For instance, Löckenhoff et al. (2009) found that a higher proportion of older adults in the population was not related to more frequent intergenerational contact with young adults. Limitations Although population aging can be considered similar in different districts of the same cluster, this does not imply that other environmental variables such as infrastructure or opportunities for leisure and engagement are similar. The district-level is the smallest regional level with available data on population aging in Germany. Nevertheless, living conditions within one district can differ to a considerable extent. In addition, individuals may differ in the degree of use of services provided by neighboring regions, and consequently may be more or less exposed to the population composition. However, this limitation speaks for less strong contextual effects and therefore even underlines the small but significant effects of differences in population aging on SPA found in this study. Generalizations of results to other than German regions should be drawn with caution. In particular, country-specific age stereotypes and country-specific legislations (e.g., welfare state) may drive some of the observed results. In addition, countries may differ in social change of SPA (Germany shows a positive pattern over time; Beyer et al., 2017) which may affect the results. Future studies should compare regions with similar population aging patterns between different countries. The clusters used reflect a share of different population aging processes. It remains unclear, if certain aspects are more strongly related to SPA than others. A high proportion of old-old people may challenge regions differently as compared to a high proportion of young-old people. In addition, other regional aspects, not reflected in the clusters, may be relevant for the SPA of the inhabitants. This study combines the investigation of social and intraindividual change in a sample with a broad age range (40–85 years). Even though FIML counteracts the effects of sample attrition, interpretations regarding intraindividual change in SPA should be made with caution. However, the combination of social and intraindividual change has the advantage of a large sample adequately representing districts and the four clusters in Germany as well as reflecting the effects of population aging over 12 years for both social and intraindividual changes in SPA. Furthermore, there might be time periods in which contextual factors have a stronger effect on individual SPA (e.g., availability of health care institutions after a health event; engagement opportunities after retirement). Continuous, closely-timed SPA assessment may help capture those time periods and allow the identification of potential intervention targets. Although indicators of wealth and health care opportunities were included as covariates, underlying mechanisms, such as leisure opportunities and economic living conditions in the districts were not fully captured in our model and may drive the random effects of the multilevel models. Future studies should additionally try to capture these district characteristics to further explain the district level variance in level and change of SPA. Conclusion As most societies face population aging, it is important to investigate its effects on individual development and aging. The current study focuses on contextual effects on self-perceptions of aging as one important psychological resource for health, well-being and longevity in later life. Results suggest that SPA do not develop independently from population aging on the district level. Rather, the pace of population aging in different regions of one single country can contribute to individual changes in positive and negative SPA. This finding complements previous cross-cultural comparisons. Districts facing fast population aging should address possible negative effects by providing attractive opportunities for families with young children, by increasing efforts to maintain and improve good infrastructure (e.g., public transport, health care, and shopping facilities), and by providing opportunities for intergenerational contact and mutual support. Additionally, health communication and public health messages that emphasize positive aspects of aging (e.g., wisdom, positive older role models) may counteract negative effects of fast population aging. Results imply that regional differences and potential contextual opportunities for aging well should be taken into account when trying to promote positive perceptions of aging. Funding This work was supported by the German Federal Ministry of Family Affairs, Senior Citizens, Women and Youth (grant number 301-6083-05/003*2) and by the German Research Foundation (DFG; grant 632451, granted to Verena Klusmann). Conflict of Interest None reported. Acknowledgments The authors thank Constantin Bolz for his support in data preparation. J. K. Wolff, A.-K. Beyer, S. Wurm, and M. Wiest are members of the Scientific Network “Images of Aging,” funded by the German Research Foundation (DFG; grant 632451, granted to Verena Klusmann). References Allport, G. W. ( 1954). The nature of prejudice . Reading, MA: Addicon Wesley. Beyer, A.-K., Wurm, S., & Wolff, J. K. ( 2017). Älter werden – Gewinn oder Verlust? Individuelle Altersbilder und Altersdiskriminierung. [Growing older – gain or loss? Individual perceptions of aging and age discrimination] In K. Mahne, J. K. Wolff, J. Simonson, & C. Tesch-Römer (Eds.), Altern im Wandel: Zwei Jahrzehnte Deutscher Alterssurvey (DEAS)  (pp. 329– 343). Wiesbaden: Springer VS. Bowen, C. E., & Skirbekk, V. ( 2013). National stereotypes of older people’s competence are related to older adults’ participation in paid and volunteer work. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 68, 974– 983. doi: 10.1093/geronb/gbt101 Google Scholar CrossRef Search ADS   Bronfenbrenner, U. ( 1977). Toward an experimental ecology of human-development. American Psychologist , 32, 513– 531. doi: 10.1037/0003-066x.32.7.513 Google Scholar CrossRef Search ADS   Bronfenbrenner, U. ( 1979). The ecology of human development: Experiments by nature and design . Cambridge, MA: Harvard University Press. Bullinger, M., & Kirchberger, I. ( 1998). Der SF-36 Fragebogen zum Gesundheitszustand. Handbuch für die deutschsprachige Fragebogenversion . Göttingen: Hogrefe Verlag. Bryant, C., Bei, B., Gilson, K. M., Komiti, A., Jackson, H., & Judd, F. ( 2016). Antecedents of attitudes to aging: A study of the roles of personality and well-being. The Gerontologist , 56, 256– 265. doi: 10.1093/geront/gnu041 Google Scholar CrossRef Search ADS PubMed  Craciun, C., Gellert, P., & Flick, U. ( 2017). Aging in precarious circumstances: Do positive views on aging make a difference? The Gerontologist , 57, 517– 528. doi: 10.1093/geront/gnv135 Google Scholar PubMed  Cagny, K. A., & Wen, M. ( 2008). Social capital and aging-related outcomes. In I. Kawachi, S. V. Subramanian, & D. Kim (Eds.), Social capital and health  (pp. 239– 258). New York: Springer. Google Scholar CrossRef Search ADS   Dannefer, D. ( 1992). On the conceptualization of context in developmental discourse: Four meanings of context and their implications. In D. L. Featherman, R. M. Lerner, & M. Perlmutter (Eds.), Life-span development and behavior  (pp. 83– 110). Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers. Donlon, M. M., Ashman, O., & Levy, B. R. ( 2005). Re-vision of older television characters: A stereotype-awareness intervention. Journal of Social Issues , 61, 307– 319. doi: 10.1111/j.1540-4560.2005.00407.x Google Scholar CrossRef Search ADS   Gerstorf, D., & Ram, N. ( 2012). Late life. In S. Krauss Whitebourne, & M. J. Sliwinski (Eds.), Wiley-Blackwell handbook of adulthood and aging  (pp. 49– 71). Oxford, UK: Wiley-Blackwell. Google Scholar CrossRef Search ADS   Giles, H., Noels, K. A., Williams, A., Ota, H., Lim, T. S., Ng, S. H., … Somera, L. ( 2003). Intergenerational communication across cultures: Young people’s perceptions of conversations with family elders, non-family elders and same-age peers. Journal of Cross-Cultural Gerontology , 18, 1– 32. doi: 10.1023/a:1024854211638 Google Scholar CrossRef Search ADS PubMed  Kim, J. E., & Moen, P. ( 2002). Retirement transitions, gender, and psychological well-being: A life-course, ecological model. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 57, P212– P222. doi: 10.1093/geronb/57.3.P212 Google Scholar CrossRef Search ADS   Klaus, D., Engstler, H., Mahne, K., Wolff, J. K., Simonson, J., Wurm, S., & Tesch-Römer, C. ( 2017). Cohort profile: The German Ageing Survey (DEAS). International Journal of Epidemiology . doi: 10.1093/ije/dyw326 Kleinspehn-Ammerlahn, A., Kotter-Grühn, D., & Smith, J. ( 2008). Self-perceptions of aging: Do subjective age and satisfaction with aging change during old age? The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 63, P377– P385. doi: 10.1093/geronb/63.6.P377 Google Scholar CrossRef Search ADS   Kornadt, A. E., & Rothermund, K. ( 2011). Contexts of aging: Assessing evaluative age stereotypes in different life domains. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 66, 547– 556. doi: 10.1093/geronb/gbr036 Google Scholar CrossRef Search ADS   Kornadt, A. E., Voss, P., & Rothermund, K. (2017). Age stereotypes and self-views revisited: Patterns of internalization and projection processes across the life span. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 72, 582–592. doi:10.1093/geronb/gbv099 Kwak, M., Ingersoll-Dayton, B., & Burgard, S. ( 2014). Receipt of care and depressive symptoms in later life: The importance of self-perceptions of aging. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 69, 325– 335. doi: 10.1093/geronb/gbt128 Google Scholar CrossRef Search ADS   Levy, B. R. ( 2008). Rigidity as a predictor of older persons’ aging stereotypes and aging self-perceptions. Social Behavior and Personality , 36, 559– 570. doi: 10.2224/sbp.2008.36.4.559 Google Scholar CrossRef Search ADS   Levy, B. ( 2009). Stereotype embodiment: A psychosocial approach to aging. Current Directions in Psychological Science , 18, 332– 336. doi: 10.1111/j.1467-8721.2009.01662.x Google Scholar CrossRef Search ADS PubMed  Li, S. C. ( 2003). Biocultural orchestration of developmental plasticity across levels: The interplay of biology and culture in shaping the mind and behavior across the life span. Psychological Bulletin , 129, 171– 194. doi: 10.1037/0033-2909.129.2.171 Google Scholar CrossRef Search ADS PubMed  Löckenhoff, C. E., De Fruyt, F., Terracciano, A., McCrae, R. R., De Bolle, M., Costa, P. T. Jr,… Yik, M. ( 2009). Perceptions of aging across 26 cultures and their culture-level associates. Psychology and Aging , 24, 941– 954. doi: 10.1037/a0016901 Google Scholar CrossRef Search ADS PubMed  Macia, E., Lahmam, A., Baali, A., Boëtsch, G., & Chapuis-Lucciani, N. ( 2009). Perception of age stereotypes and self-perception of aging: A comparison of French and Moroccan populations. Journal of Cross-Cultural Gerontology , 24, 391– 410. doi: 10.1007/s10823-009-9103-0 Google Scholar CrossRef Search ADS PubMed  McCann, R. M., & Keaton, S. A. ( 2013). A cross cultural investigation of age stereotypes and communication perceptions of older and younger workers in the USA and Thailand. Educational Gerontology , 39, 326– 341. doi: 10.1080/03601277.2012.700822 Google Scholar CrossRef Search ADS   Menning, S., Nowossadeck, E., & Maretzke, S. ( 2010). Regionale Aspekte der demografischen Alterung. [Regional aspects of demographic aging]. Report Altersdaten , 1–2, 1– 56. Retrieved from http://nbn-resolving.de/urn:nbn:de:0168-ssoar-369378. Moor, C., Zimprich, D., Schmitt, M., & Kliegel, M. ( 2006). Personality, aging self-perceptions, and subjective health: A mediation model. International Journal of Aging & Human Development , 63, 241– 257. doi: 10.2190/AKRY-UM4K-PB1V-PBHF Google Scholar CrossRef Search ADS PubMed  Mun, E. Y., von Eye, A., & White, H. R. ( 2009). An SEM approach for the evaluation of intervention effects using pre-post-post designs. Structural Equation Modeling: A Multidisciplinary Journal , 16, 315– 337. doi: 10.1080/10705510902751358 Google Scholar CrossRef Search ADS PubMed  North, M. S., & Fiske, S. T. ( 2015). Modern attitudes toward older adults in the aging world: A cross-cultural meta-analysis. Psychological Bulletin , 141, 993– 1021. doi: 10.1037/a0039469 Google Scholar CrossRef Search ADS PubMed  Pettigrew, T. F., & Tropp, L. R. ( 2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology , 90, 751– 783. doi: 10.1037/0022-3514.90.5.751 Google Scholar CrossRef Search ADS PubMed  Putnam, R. D. ( 2000). Bowling alone . New York: Simon & Schuster. Rentfrow, P. J., & Jokela, M. ( 2016). Geographical psychology: The spatial organization of psychological phenomena. Current Directions in Psychological Science , 25, 393– 398. doi: 10.1177/0963721416658446 Google Scholar CrossRef Search ADS   Sargent-Cox, K. A., Anstey, K. J., & Luszcz, M. A. ( 2012). Change in health and self-perceptions of aging over 16 years: The role of psychological resources. Health Psychology: Official Journal of the Division of Health Psychology, American Psychological Association , 31, 423– 432. doi: 10.1037/a0027464 Google Scholar CrossRef Search ADS PubMed  Schafer, J. L., & Graham, J. W. ( 2002). Missing data: Our view of the state of the art. Psychological Methods , 7, 147– 177. doi: 10.1037/1082-989x.7.2.147 Google Scholar CrossRef Search ADS PubMed  Silverstein, M., Parrott, T. M., Angelelli, J. J., & Cook, F. L. ( 2000). Solidarity and tension between age groups in the United States: Challenge for an aging America in the 21st century. International Journal of Social Welfare , 9, 270– 284. doi: 10.1111/1468–2397.00139 Google Scholar CrossRef Search ADS   Steverink, N., Westerhof, G. J., Bode, C., & Dittmann-Kohli, F. ( 2001). The personal experience of aging, individual resources, and subjective well-being. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 56, P364– P373. doi: 10.1093/geronb/56.6.P364 Google Scholar CrossRef Search ADS   Thompson, E. H. Jr, & Weaver, A. J. ( 2016). Making connections: The legacy of an intergenerational program. The Gerontologist , 56, 909– 918. doi: 10.1093/geront/gnv064 Google Scholar CrossRef Search ADS PubMed  UNESCO.( 2012). International Standard Classification of Education ISCED 2011 . Montreal: UNESCO Institute for Statistics. Vitman, A., Iecovich, E., & Alfasi, N. ( 2014). Ageism and social integration of older adults in their neighborhoods in Israel. The Gerontologist , 54, 177– 189. doi: 10.1093/geront/gnt008 Google Scholar CrossRef Search ADS PubMed  Westerhof, G. J., Miche, M., Brothers, A. F., Barrett, A. E., Diehl, M., Montepare, J. M., … Wurm, S. ( 2014). The influence of subjective aging on health and longevity: A meta-analysis of longitudinal data. Psychology and Aging , 29, 793– 802. doi: 10.1037/a0038016 Google Scholar CrossRef Search ADS PubMed  Wurm, S., Tesch-Römer, C., & Tomasik, M. J. ( 2007). Longitudinal findings on aging-related cognitions, control beliefs, and health in later life. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 62, P156– P164. doi: 10.1093/geronb/62.3.p156 Google Scholar CrossRef Search ADS   Wurm, S., Wolff, J. K., & Schüz, B. ( 2014). Primary care supply moderates the impact of diseases on self-perceptions of aging. Psychology and Aging , 29, 351– 358. doi: 10.1037/a0036248 Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Gerontologist Oxford University Press

Regional Impact of Population Aging on Changes in Individual Self-perceptions of Aging: Findings From the German Ageing Survey

Loading next page...
 
/lp/oxford-university-press/regional-impact-of-population-aging-on-changes-in-individual-self-UZgt1WQJm0

References (37)

Publisher
Oxford University Press
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0016-9013
eISSN
1758-5341
DOI
10.1093/geront/gnx127
pmid
28958001
Publisher site
See Article on Publisher Site

Abstract

Abstract Background and Objectives The importance of self-perceptions of aging (SPA) for health and longevity is well documented. Comparably little is known about factors that contribute to SPA. Besides individual factors, the context a person lives in may shape SPA. Research has so far focused on country-level differences in age stereotypes, indicating that rapid population aging accompanies more negative age stereotypes. The present study expands previous research by investigating the impact of district-specific population aging within one country on different facets of SPA. Research Design and Methods Based on a large representative survey in Germany, the study investigates changes in SPA as ongoing development as well as the SPA of physical loss over a 12-year period in adults aged 40+. The study uses several indicators of population aging (e.g., population development, average age, greying index), to identify four clusters differing in their pace of population aging. Based on three-level latent change models, these clusters were compared in their impact on changes in SPA. Results Compared to districts with an average rate of population aging, the study shows that persons living in regions with a fast population aging rate (C1) hold more negative SPA in both facets (ps = .01). Districts with slow population aging (C2) have significantly higher SPA ongoing development (p = .03). Discussion and Implications The study underlines the importance for regional differences in population aging on the development of SPA. In particular, societies should be aware that fast population aging may result in more negative SPA. Attitudes and perception toward aging/aged, Demography, context factors/contextual influences, Successful aging A growing body of research underlines the importance of investigating contextual influences on psychological phenomena such as attitudes and values (Rentfrow & Jokela, 2016). Gerstorf and Ram (2012) collected empirical evidence on the effects of contextual factors on developmental trajectories in later life. The authors state that four components of context may be linked to individual development: economic characteristics (e.g., level of poverty or rate of unemployment), service characteristics (e.g., availability and accessibility of health services), social characteristics (e.g., social ties or perceived discrimination), and physical characteristics (e.g., features of the built environment or accessibility). Aging well is thus shaped by individual resources and constraints as well as contextual factors on various levels (e.g., country, region, and neighborhood). An important indicator of aging well is the individual view on one’s own aging process. Numerous studies have illustrated the far-reaching importance of self-perceptions of aging (SPA) for health, well-being, and longevity (Westerhof et al., 2014). However, little is known about which factors, besides individual characteristics, may shape SPA. Considering the speed of current demographic change and population aging in most industrialized countries, their impact on SPA should clearly be assessed. Thus, the current study investigates the effect differences in the pace of population aging on SPA have regionally, over 12 years. Determinants of SPA In accordance with stereotype embodiment theory (Levy, 2009) and supported by empirical findings (e.g., Kornadt, Voss, & Rothermund, 2017), age stereotypes are learned and internalized during youth, accumulate over the life span, and become SPA when they are directed at oneself and the own aging process with increasing age. Changes in SPA can be investigated from two perspectives: intraindividual and social change. A recent study on social change in SPA in the German population between 1996 and 2014 showed that SPA have improved in persons aged 40–85 years (Beyer, Wurm, & Wolff, 2017). Longitudinal studies on intraindividual changes in SPA with age are rare (e.g., Kleinspehn-Ammerlahn, Kotter-Grühn, & Smith, 2008) but show, in line with cross-sectional age group comparisons, that older individuals report more negative SPA than younger individuals (e.g., Wurm, Wolff, & Schüz, 2014). Intraindividual and social change of age stereotypes and SPA can be influenced by individual factors such as socioeconomic variables (e.g., Kleinspehn-Ammerlahn et al., 2008), health status (e.g., Sargent-Cox, Anstey, & Luszcz, 2012; Wurm, Tesch-Römer, & Tomasik, 2007), personality (e.g., Bryant et al., 2016; Levy, 2008; Moor, Zimprich, Schmitt, & Kliegel, 2006), and own and vicarious aging experiences, such as receipt of care (e.g., Kwak, Ingersoll-Dayton, & Burgard, 2014). Contextual influences may be defined as “(…) the totality of the diverse range of phenomena, events, and forces that exist outside the developing individual, including its sociocultural and physical aspects” (Dannefer, 1992, p. 84). Additionally, the environment a person lives in changes over time (Bronfenbrenner, 1977), and, according to the ecological systems theory (Bronfenbrenner, 1979) and the cross-level dynamic biocultural coconstructive framework of development (Li, 2003), contextual variables influence behavior and attitudes of individuals. Empirically, contextual variables such as the portrayal of age and older adults in the media (e.g., Donlon, Ashman, & Levy, 2005), age-graded social regulations such as retirement (e.g., Kim & Moen, 2002), or the societal roles of older adults (e.g., Bowen & Skirbekk, 2013) have been shown to play a role in shaping age stereotypes and their internalization into SPA. However, even though many western societies face demographic changes, characteristics of population aging at the regional level have been rarely examined as contextual influences on SPA. Population Aging and SPA While the effect of population aging on SPA has not been examined so far, first evidence suggests that population aging affects age stereotypes and attitudes towards aging. A comparison of eastern and western cultures showed overall more favorable age stereotypes in western cultures (e.g., Giles et al., 2003; McCann & Keaton, 2013), which seems to be accounted for by differences in population aging (see meta-analysis by North & Fiske, 2015 for details). North and Fiske (2015) showed that individuals living in countries with rapid population aging, which is true for most eastern countries, have more negative age stereotypes. In a similar vein, Löckenhoff et al. (2009) compared aging perceptions of mostly younger adults in 26 countries and found that countries with a higher proportion of inhabitants aged 65+ had more negative perceptions of aging. These findings, however, contradict the assumption of the contact hypothesis (cf. Allport, 1954; Pettigrew & Tropp, 2006), which suggests that having more contact with a group of people will result in reduced prejudices towards that group. That is, individuals living in contexts with comparatively more older adults should have fewer prejudices toward older adults, which could, in turn, be associated with decreased negative age stereotypes. However, these studies focused on age stereotypes or attitudes toward aging and did not investigate how an individual perceives his or her own aging process. Regarding the relation between population aging and SPA, there are good arguments for both more positive and more negative SPA due to population aging: On the one hand, increased contact with older adults due to living in an area with a larger older adult population could reduce negative SPA as population aging may enhance the likelihood of experiencing healthy older adults, who actively participate in society. On the other hand, population aging can impose threats and challenges on a society. For example, anticipated problems in distributive justice of health care and pension systems, increased care needs of old-old individuals, or the financial burden of younger family members or the younger generation in general may result in more negative age stereotypes and SPA. In particular, rapid population aging may hamper adaptation to these changes. Silverstein, Parrott, Angelelli, and Cook (2000) argue that a greater proportion of older adults may result in intergenerational conflict over limited resources. Cagny and Wen (2008) state that age structure drives the demand for services (e.g., health clinics) and infrastructure (e.g., sidewalks) as well as fostering expectations about social roles (e.g., informal monitoring of children) that may affect social contact and cohesion in neighborhoods. Putnam (2000) found that a high number of aged residents is associated with a more active neighborhood watch, better social services, and, in general, a community more engaged in civic affairs. Furthermore, social characteristics on a regional level affect individuals via informal social control (e.g., monitoring, expectations for action, availability of role models; cf. Cagny & Wen, 2008; Gerstorf & Ram, 2012). These changes in infrastructure and norms may provide more opportunities for intergenerational contact resulting in stronger social cohesion. For example, increased intergenerational contact due to a community program was already shown to result in more positive images of aging in fourth grade students (even though no effect was found for negative images of aging; Thompson & Weaver, 2016). A country level perspective, however, ignores heterogeneity in population aging within a country. Aspects of population aging, such as increases in the proportion of elderly population, do not happen in the same manner in all geographical regions of one country (Menning, Nowossadeck, & Maretzke, 2010). In addition, regional patterns of population aging affect more directly the immediate environment where people live as well as the perception of older people and may, therefore, have stronger effects on SPA. In Germany, districts are responsible for various infrastructural aspects, such as establishing economic infrastructures, planning for housing areas, providing health care services, and offering opportunities for volunteering. Previous research has largely neglected regional differences in population aging, age stereotypes, and SPA, even though there are some hints of regional differences in age stereotypes for example between urban and rural areas (favoring rural areas; Macia, Lahmam, Baali, Boëtsch, & Chapuis-Lucciani, 2009). Thus, the analysis of population aging at the district-level provides additional information to previous studies on country-level aging: First, the heterogeneity of population aging within one country is captured, and, second, this perspective takes into account that age stereotypes and SPA emerge under specific regional circumstances of social, economic, and cultural life. The Present Study No study has thus far examined the impact of population aging on SPA. The present study therefore explores the question whether district level differences in population aging in Germany explain changes in SPA. Population aging is an interaction of different demographic factors and a complex phenomenon that cannot be captured sufficiently with single indicators, such as percentage of older adults (as used in most of the previous studies). Therefore, we decided to use four clusters reflecting differences in indicators of population aging (e.g., average age; proportion of very old individuals [80+] at district level) were used in the analyses (see Method section for details; Menning et al., 2010). The clusters should explain variability in change in two different facets of SPA: the experience of aging as both ongoing development and physical loss. These two facets capture two important dimensions of the multidimensional construct of SPA (cf. Kornadt & Rothermund, 2011; Steverink, Westerhof, Bode, & Dittmann-Kohli, 2001; Wurm et al., 2007). Ongoing development describes the belief that aging is accompanied with new opportunities and one is still making new plans. Chances for ongoing development are affected by contextual factors such as opportunities for leisure or other activities. The second facet captures the common belief that aging is strongly associated with physical loss. This loss-related view might also be prone to contextual influences, for example, in terms of environments equipped for older and functional disabled persons. This study examines population aging at the district level and changes in two facets of SPA across 12 years using data from a large representative survey of adults aged 40 years and older in Germany to answer the research question: Do different regional patterns of population aging influence changes in SPA? Design and Methods Sample Data come from two measurement points (1996, 2008) of the German Ageing Survey (DEutscher AltersSurvey [DEAS]; Klaus et al., 2017). DEAS is a national representative cohort-sequential study of adults aged 40 years and older. Cross-sectional baseline samples have been drawn every 6 years since 1996 and are stratified by age, gender, and region (Eastern/Western Germany). The sampling procedure of the DEAS starts with randomly drawn municipalities, in which stratified samples are randomly drawn from registry offices. Thereby, the sample is representative of the German population, with only a subsample of all 413 German districts in the data. At each measurement point, participants took part in a 90-min interview and filled in a paper-pencil questionnaire. The current study combines cross-sectional and longitudinal information from two measurement points (t1: 1996 and t2: 2008; see Figure 1) to assess a combination of social change (between baseline samples of 1996 and 2008) and intraindividual change (between t1 [1996] and t2 [2008]) in SPA. The 1996 sample data represents a baseline at t1 (n = 4,838). The 2008 sample includes both baseline data from additional participants (n = 6,205) as well as the longitudinal data from the 1996 participants. Due to political reforms that led to changes in the number of districts in Germany, 155 persons had to be excluded because they could not be clearly assigned to one of the clusters (n1996 = 32; n2008 = 123). As SPA were assessed in the paper-pencil questionnaire, only participants who completed the questionnaire were included in the analyses: Of the remaining sample from 1996, n = 797 persons did not fill-in the paper-pencil questionnaire in 1996 and n = 148 in 2008. Of the remaining sample in 2008, n = 1,699 participants had to be excluded due to missing paper-pencil questionnaires. Additionally, we excluded all persons who moved between districts between measurement points (n = 45 of the 1996 sample). The analyzed sample in 1996 includes 3,816 persons. The analyzed sample in 2008 consists of 5,122 persons, including information from the 2008 baseline sample (n = 4,383) and longitudinal information from the 1996 sample (n = 739). Thus, a total sample of n = 8,199 was analyzed. Of the 413 districts in Germany, 207 are represented in the data. On average, 40 persons lived in each district. Figure 1. View largeDownload slide Flow-chart of DEAS participants used in the study; p&p = paper & pencil questionnaire; B1996 = baseline sample 1996; L2008 = longitudinal sample of baseline participants 1996; B2008 = baseline sample 2008. DEAS = DEutscher AltersSurvey. Figure 1. View largeDownload slide Flow-chart of DEAS participants used in the study; p&p = paper & pencil questionnaire; B1996 = baseline sample 1996; L2008 = longitudinal sample of baseline participants 1996; B2008 = baseline sample 2008. DEAS = DEutscher AltersSurvey. Measures Demographic Clusters Four clusters differing in their pace of population aging were identified by a cluster analysis (Menning et al., 2010) aimed at describing population aging in all German districts based on the following indicators: (a) population development per 1,000 inhabitants between 1995 and 2008, (b) average age of population in 2008, (c) percentage of population aged 65+ in 1995, (d) percentage of population aged 65+ in 2008, (e) greying index (old-old population, aged 80+ years, in relation to young-old, aged 65–79 years) in 2008, and (f) balance of births and deaths per 1,000 inhabitants in 2008. The average values of these indicators from each cluster for all German districts are summarized in Table 1 and described below. Table 1. Cluster Indicators of Population Aging in Germany (average values, 413 districts in 2008) Indicators  Cluster 1 fast aging; 2008: highest average age/low proportion of 80+  Cluster 2 slow aging; 2008: lowest average age/low proportion of 80+  Cluster 3 average aging; 2008: high average age/high proportion of 80+  Cluster 4 average aging; 2008: average age/average proportion of 80+  Population development per 1,000 inhabitants, between 1995 and 2008 (balance)  −119.7  +70.5  −31.3  +17.1  Average age of population in 2008 (years)  45.8  42.0  44.3  43.0  Share of population aged 65 years and older in 2008 (%)  23.5  18.6  22.4  20.3  Share of population aged 65 years and older in 1995 (%)  14.9  13.8  18.4  15.9  Greying index (very old population aged 80 years and older in relation to the young elderly aged from 65 to 79 years) in 2008 (ratio)  27.6  30.3  37.1  33.4  Balance of births and deaths per 1,000 inhabitants in 2008 (birth surplus)  −4.5  −0.4  −4.4  −2.2  Indicators  Cluster 1 fast aging; 2008: highest average age/low proportion of 80+  Cluster 2 slow aging; 2008: lowest average age/low proportion of 80+  Cluster 3 average aging; 2008: high average age/high proportion of 80+  Cluster 4 average aging; 2008: average age/average proportion of 80+  Population development per 1,000 inhabitants, between 1995 and 2008 (balance)  −119.7  +70.5  −31.3  +17.1  Average age of population in 2008 (years)  45.8  42.0  44.3  43.0  Share of population aged 65 years and older in 2008 (%)  23.5  18.6  22.4  20.3  Share of population aged 65 years and older in 1995 (%)  14.9  13.8  18.4  15.9  Greying index (very old population aged 80 years and older in relation to the young elderly aged from 65 to 79 years) in 2008 (ratio)  27.6  30.3  37.1  33.4  Balance of births and deaths per 1,000 inhabitants in 2008 (birth surplus)  −4.5  −0.4  −4.4  −2.2  View Large Cluster 1 (C1; fast aging; 2008: highest average age/low proportion of 80+) Cluster 1 comprises districts with a rapidly aging, shrinking population. All districts in this cluster are located in Eastern Germany. In the past, these districts had a relatively young population structure. Since 1995 they have been confronted with population decline and an accelerated aging process. The average age in 2008 is the highest of all clusters. The percentage of people aged 65+ is above average, although the percentage of people aged 80+ is the smallest among all clusters. In this study, 42 districts with 2,084 participants belong to C1. Cluster 2 (C2; slow aging; 2008: lowest average age/low proportion of 80+) The districts of this cluster are mostly situated in urban agglomerations and are characterized by a slowly aging, growing population. The population has been increasing since 1995, mostly due to internal migration. The average age in 2008 is lower than the average for Germany. Moreover, the percentages of people aged 65+ or 80+ are low in comparison to the other clusters. The districts of C2 have stable and growing populations. Demographic change is occurring at a much slower pace. In this study, 53 districts with 2,266 participants belong to C2. Cluster 3 (C3; average aging; 2008: high average age/high proportion of 80+) This cluster comprises districts with a slowly aging, shrinking population. All districts in this cluster are located in Western Germany. Population aging in this cluster has progressed considerably: In contrast to C1 it started with a much older population in 1995; since then, the pace of population aging has been slower than in C1. The percentage of the population aged 65+ is higher than on average. The greying index is the highest of all clusters in 2008. In this study, 31 districts and 849 participants belong to C3. Cluster 4 (C4; average aging; 2008: average age/average proportion of 80+) Cluster 4 consists of districts with an average rate of population aging. The vast majority of districts in this cluster are situated in Western Germany. They are characterized by a medium pace of population aging and minimal population growth since 1995. Most of the indicators of population aging are close to the average for Germany. In this study, 81 districts and 3,000 participants belong to C4. For the analyses, the clusters were dummy coded using C4 as reference category as it represents average population aging in Germany. SPA Two scales of SPA were assessed in 1996 and 2008 with subscales for the aging-related cognition of ongoing development and physical loss (AgeCog-Scales; Steverink et al., 2001; Wurm et al., 2007). The SPA ongoing development consists of four items and refers to the positive view that aging is seen as a time of continuous personal development (e.g., “Aging means to me that I can still learn new things,” “Aging means to me that my capabilities are increasing”; Cronbach’s α1996 = .91; Cronbach’s α2008 = .93). The SPA physical losses consists of four items and addresses the view that aging is accompanied by physical decline (e.g., “Aging means to me that I am less healthy,” “Aging means to me that I am less energetic and fit”; Cronbach’s α1996 = .91; Cronbach’s α2008 = .91). Responses could range from 1 (strongly disagree) to 4 (strongly agree). For each scale, the four items were averaged to obtain a score, with higher values indicating greater agreement. That is, a higher SPA score for ongoing development indicates a more positive view, and a higher SPA score for physical losses indicates a more negative view. Covariates (individual) Covariates on the individual level were age, gender, and region (Eastern/Western Germany) as the DEAS sample is stratified by theses variables and to account for age-related differences in SPA. As more healthy and highly educated people are known to report more positive SPA, analyses were also controlled for education. This was assessed using the International Standard Classification of Education (ISCED with three levels; UNESCO, 2012). The three levels are: low education (9 years of school education at most), medium education (secondary school), and high education (qualifying for university admission). Additionally, the health status of the participants was controlled for using both self-reported numbers of physical diseases (up to 11 diseases such as cardiovascular diseases, back or joint diseases, respiratory diseases) and functional health (physical functioning subscale of the SF-36 questionaire; Bullinger & Kirchberger, 1998). The latter was coded according to the manual, with values ranging from 1 to 100 and higher values indicating better functional health. Values of these covariates are based on the first assessment point of each participant. Covariates (district) The variables included as covariates are indicators of the wealth and health care opportunities of the district, which have been shown to be related to SPA and psychological resources (Wurm et al., 2014). On a district level, all analyses were controlled for using the percentage of unemployed older workers (55–65 years) from all unemployed persons (M [SD] = 13.56 [2.31]), primary care supply (number of general practitioners per 100,000 inhabitants; M [SD] = 673.30 [179.51]), gross domestic product (GDP in 1000 Euro per capita; M [SD] = 28.66 [12.37]), and population density (number of residents per km2; M [SD] = 749.82 [967.97]) in 2008. In addition, the analyses were controlled for using the district level average remaining life expectancy for 60-year-old persons between 2007 and 2009 (M [SD] = 23.21 [0.74]). Primary care supply, GDP, and population density were divided by 100 before including them in the analyses to facilitate the estimation of the models. Data Analyses The data has a three-level structure with time on Level 1, individuals on Level 2, and districts on Level 3. Therefore, a three-level latent change model was used to analyze the data in Mplus (adapted from Mun, von Eye, & White, 2009). Within-person change in SPA across 12 years was modeled, allowing for individual and district-level differences in this change as well as in intercept. Change in SPA ongoing development and SPA physical losses were estimated in two separate models. The intercept was modeled to represent levels at 2008 allowing for testing level differences in SPA in 2008. Intercepts and change scores were predicted by the cluster dummies (C4 as reference) and all covariates. Estimation was conducted with full information maximum likelihood estimation (FIML) to account for longitudinal drop-out in the sample (cf. Schafer & Graham, 2002). Significance level was set to .05. All Level 2 variables were group-mean centered. All Level 3 variables were grand-mean centered. Covariates were allowed to covary on individual and district level. To ensure model identification, the residual variances of SPA ongoing development and SPA physical losses were set zero for 1996, and the intercept and change scores were not allowed to covary on individual level. Results Descriptive Statistics The average age at first measurement point was 61.06 years (SD = 12.05, range: 40–85 years) and 48.70% were female. Of all participants, 12.96% had low, 56.04% medium, and 31.00% high-level education. On average, the score on the physical functioning subscale of the SF-36 was M (SD) = 83.99 (22.37) and participants reported 2.41 diseases (SD = 1.87). SPA Ongoing Development The intraclass correlation coefficient (ICC: the proportion of variance on district level to total variance) for SPA ongoing development was .05 and .06 in 1996 and 2008, respectively. The results of the multilevel model are summarized in Table 2. Table 2. Unstandardized Parameters of Multilevel Latent Change Model for SPA Ongoing Development using DEAS Data From 1996 to 2008 Parameter  Estimate (SE)  p value  Fixed Effects (Level 1 Time)   Intercept (level 2008)  2.95 (0.03)  <.001   Change (1996–2008)  −0.03 (0.03)  .30  Fixed Effects (Level 2 Individual)  Intercept on       Age  −0.01 (0.001)  <.001   Gender  0.05 (0.02)  .001   Eastern/Western Germany  −0.12 (0.06)  .03   Education  0.13 (0.01)  <.001   Functional Health  0.01 (0.001)  <.001   Number of Diseases  −0.03 (0.01)  <.001  Change on       Age  −0.01 (0.001)  <.001   Gender  0.02 (0.02)  .42   Eastern/Western Germany  0.03 (0.05)  .51   Education  0.003 (0.02)  .89   Functional Health  −0.003 (0.001)  <.001   Number of Diseases  0.01 (0.01)  .38  Fixed Effects (Level 3 District)  Intercept on       C1 vs C4  −0.10 (0.04)  .01   C2 vs C4  0.03 (0.04)  .50   C3 vs C4  0.01 (0.04)  .88   Unemployment Rate 2008  −0.001 (0.01)  .89   Primary Care Supply 2008  −0.01 (0.01)  .13   GDP 2008  −0.08 (0.13)  .54   Remaining Life Expectancy 2008  0.01 (0.02)  .59   Population Density 2008  −0.002 (0.003)  .56  Change on       C1 vs C4  −0.04 (0.05)  .38   C2 vs C4  −0.10 (0.05)  .03   C3 vs C4  −0.04 (0.05)  .46   Unemployment Rate 2008  −0.002 (0.01)  .75   Primary Care Supply 2008  0.01 (0.01)  .75   GDP 2008  −0.01 (0.15)  .94   Remaining Life Expectancy 2008  −0.02 (0.03)  .49   Population Density 2008  0.002 (0.003)  .55  Random Effects (Level 2 Individual)   Intercept  0.13 (0.02)  <.001   Change  0.24 (0.02)  <.001  Random Effects (Level 3 District)   Intercept  0.02 (0.01)  <.001   Change  0.01 (0.01)  .11  Random Effects (Level 1 Time)   Residual  0.17 (0.02)  <.001  Parameter  Estimate (SE)  p value  Fixed Effects (Level 1 Time)   Intercept (level 2008)  2.95 (0.03)  <.001   Change (1996–2008)  −0.03 (0.03)  .30  Fixed Effects (Level 2 Individual)  Intercept on       Age  −0.01 (0.001)  <.001   Gender  0.05 (0.02)  .001   Eastern/Western Germany  −0.12 (0.06)  .03   Education  0.13 (0.01)  <.001   Functional Health  0.01 (0.001)  <.001   Number of Diseases  −0.03 (0.01)  <.001  Change on       Age  −0.01 (0.001)  <.001   Gender  0.02 (0.02)  .42   Eastern/Western Germany  0.03 (0.05)  .51   Education  0.003 (0.02)  .89   Functional Health  −0.003 (0.001)  <.001   Number of Diseases  0.01 (0.01)  .38  Fixed Effects (Level 3 District)  Intercept on       C1 vs C4  −0.10 (0.04)  .01   C2 vs C4  0.03 (0.04)  .50   C3 vs C4  0.01 (0.04)  .88   Unemployment Rate 2008  −0.001 (0.01)  .89   Primary Care Supply 2008  −0.01 (0.01)  .13   GDP 2008  −0.08 (0.13)  .54   Remaining Life Expectancy 2008  0.01 (0.02)  .59   Population Density 2008  −0.002 (0.003)  .56  Change on       C1 vs C4  −0.04 (0.05)  .38   C2 vs C4  −0.10 (0.05)  .03   C3 vs C4  −0.04 (0.05)  .46   Unemployment Rate 2008  −0.002 (0.01)  .75   Primary Care Supply 2008  0.01 (0.01)  .75   GDP 2008  −0.01 (0.15)  .94   Remaining Life Expectancy 2008  −0.02 (0.03)  .49   Population Density 2008  0.002 (0.003)  .55  Random Effects (Level 2 Individual)   Intercept  0.13 (0.02)  <.001   Change  0.24 (0.02)  <.001  Random Effects (Level 3 District)   Intercept  0.02 (0.01)  <.001   Change  0.01 (0.01)  .11  Random Effects (Level 1 Time)   Residual  0.17 (0.02)  <.001  Note: DEAS = DEutscher AltersSurvey; GDP = Gross domestic product; SPA = Self-perceptions of aging. View Large In 2008, the average value of SPA ongoing development was 2.95 in the reference group (C4). Individuals living in C1 reported a lower level of SPA ongoing development than C4 (bC1 = −0.10, p = .01). C2 as well as C3 had a similar level compared to C4 because there was no significant difference in the intercept (bC2 = 0.03, p = .50; bC3 = 0.01, p = .88). The cluster variables explained 4.55% of the district-level variance in the intercept. Regarding changes from 1996 to 2008 (see Figure 2), the average value of SPA ongoing development remained stable over time in the reference group C4 (change = −0.03, p = .30). C1 and C3 did not differ from C4 in changes of SPA ongoing development (bC1 = −0.04, p = .38; bC3 = −0.04, p = .46), meaning SPA also remained stable over time in C1 and C3. Change over time differed between C2 and the reference group C4, as indicated by the significant effects of C2 on the change score (bC2 = −0.10, p = .03). That is, individuals living in C2 reported more positive SPA in 2008 than in 1996. The cluster variables explained 9.09% of the district-level slope variance. Figure 2. View largeDownload slide Change in SPA ongoing development from 1996 to 2008 in clusters of population aging using DEAS data. DEAS = DEutscher AltersSurvey; SPA = Self-perceptions of aging. Figure 2. View largeDownload slide Change in SPA ongoing development from 1996 to 2008 in clusters of population aging using DEAS data. DEAS = DEutscher AltersSurvey; SPA = Self-perceptions of aging. In sum, C1 showed the lowest levels of SPA ongoing development, meaning more negative SPA were reported in C1 as compared to the other clusters in 2008. Only C2 showed a significant increase in SPA ongoing development over time. The other clusters remained stable. SPA Physical Losses Results of the multilevel model for SPA physical losses are shown in Table 3. The ICC was .02 for SPA physical losses in 1996 and .04 in 2008. Table 3. Unstandardized Parameters of Multilevel Latent Change Model for SPA Physical Losses using DEAS Data From 1996 to 2008 Parameter  Estimate (SE)  p value  Fixed Effects (Level 1 Time)   Intercept (level 2008)  2.65 (0.03)  <.001   Change (1996–2008)  −0.004 (0.03)  .90  Fixed Effects (Level 2 Individual)  Intercept on       Age  0.001 (0.001)  .11   Gender  −0.04 (0.02)  .02   Eastern/Western Germany  0.11 (0.07)  .10   Education  −0.04 (0.01)  .01   Functional Health  −0.01 (0.000)  <.001   Number of Diseases  0.06 (0.01)  <.001  Change on       Age  0.01 (0.001)  <.001   Gender  0.03 (0.03)  .34   Eastern/Western Germany  −0.12 (0.06)  .06   Education  −0.02 (0.02)  .47   Functional Health  0.004 (0.001)  <.001   Number of Diseases  0.03 (0.01)  <.001  Fixed Effects (Level 3 District)  Intercept on       C1 vs C4  0.09 (0.04)  .01   C2 vs C4  0.004 (0.03)  .89   C3 vs C4  0.02 (0.04)  .58   Unemployment Rate 2008  −0.01 (0.01)  .18   Primary Care Supply 2008  −0.002 (0.01)  .82   GDP 2008  −0.01 (0.11)  .91   Remaining Life Expectancy 2008  0.01 (0.02)  .56   Population Density 2008  −0.003 (0.003)  .25  Change on       C1 vs C4  −0.04 (0.05)  .44   C2 vs C4  0.05 (0.04)  .22   C3 vs C4  −0.06 (0.06)  .28   Unemployment Rate 2008  −0.01 (0.01)  .53   Primary Care Supply 2008  0.001 (0.01)  .96   GDP 2008  0.04 (0.16)  .82   Remaining Life Expectancy 2008  −0.03 (0.03)  .31   Population Density 2008  0.000 (0.003)  .99  Random Effects (Level 2 Individual)   Intercept  0.11 (0.01)  <.001   Change  0.23 (0.01)  <.001  Random Effects (Level 3 District)   Intercept  0.01 (0.003)  <.001   Change  0.02 (0.01)  .004  Random Effects (Level 1 Time)   Residual  0.14 (0.01)  <.001  Parameter  Estimate (SE)  p value  Fixed Effects (Level 1 Time)   Intercept (level 2008)  2.65 (0.03)  <.001   Change (1996–2008)  −0.004 (0.03)  .90  Fixed Effects (Level 2 Individual)  Intercept on       Age  0.001 (0.001)  .11   Gender  −0.04 (0.02)  .02   Eastern/Western Germany  0.11 (0.07)  .10   Education  −0.04 (0.01)  .01   Functional Health  −0.01 (0.000)  <.001   Number of Diseases  0.06 (0.01)  <.001  Change on       Age  0.01 (0.001)  <.001   Gender  0.03 (0.03)  .34   Eastern/Western Germany  −0.12 (0.06)  .06   Education  −0.02 (0.02)  .47   Functional Health  0.004 (0.001)  <.001   Number of Diseases  0.03 (0.01)  <.001  Fixed Effects (Level 3 District)  Intercept on       C1 vs C4  0.09 (0.04)  .01   C2 vs C4  0.004 (0.03)  .89   C3 vs C4  0.02 (0.04)  .58   Unemployment Rate 2008  −0.01 (0.01)  .18   Primary Care Supply 2008  −0.002 (0.01)  .82   GDP 2008  −0.01 (0.11)  .91   Remaining Life Expectancy 2008  0.01 (0.02)  .56   Population Density 2008  −0.003 (0.003)  .25  Change on       C1 vs C4  −0.04 (0.05)  .44   C2 vs C4  0.05 (0.04)  .22   C3 vs C4  −0.06 (0.06)  .28   Unemployment Rate 2008  −0.01 (0.01)  .53   Primary Care Supply 2008  0.001 (0.01)  .96   GDP 2008  0.04 (0.16)  .82   Remaining Life Expectancy 2008  −0.03 (0.03)  .31   Population Density 2008  0.000 (0.003)  .99  Random Effects (Level 2 Individual)   Intercept  0.11 (0.01)  <.001   Change  0.23 (0.01)  <.001  Random Effects (Level 3 District)   Intercept  0.01 (0.003)  <.001   Change  0.02 (0.01)  .004  Random Effects (Level 1 Time)   Residual  0.14 (0.01)  <.001  Note: DEAS = DEutscher AltersSurvey; GDP = Gross domestic product; SPA = Self-perceptions of aging. View Large In 2008, the average value of SPA physical loss was 2.65 in the reference group (C4). C2 and C3 did not differ in intercept from C4 (bC2 = 0.004, p = .89; bC3 = 0.02, p = .58), which means that individuals living in C2 and C3 reported similar levels for SPA physical losses, compared to individuals living in C4. However, C1 had higher levels of SPA physical losses in 2008 as compared to C4, as indicated by a significant effect of C1 on the intercept (b = 0.09, p = .01). In total, 7.69% of the intercept variance was explained by the cluster variables. Regarding changes from 1996 to 2008 (see Figure 3), the average value of SPA physical losses remained stable over time in the reference group (change = −0.004, p = .90). Figure 3. View largeDownload slide Change in SPA physical losses from 1996 to 2008 in clusters of population aging using DEAS data. DEAS = DEutscher AltersSurvey; SPA = Self-perceptions of aging. Figure 3. View largeDownload slide Change in SPA physical losses from 1996 to 2008 in clusters of population aging using DEAS data. DEAS = DEutscher AltersSurvey; SPA = Self-perceptions of aging. None of the other clusters differed in change over time from C4 (bC1 = −0.04, p = .44; bC2 = 0.05, p = .22; bC3 = −0.06, p = .28), indicating no significant change in SPA physical losses between 1996 and 2008 in C1, C2, or C3. In sum, C1 showed higher levels of SPA physical losses, which indicates more negative SPA in C1 as compared to the other clusters. SPA physical losses did not change over time in all clusters. Discussion This study is one of the first to investigate regional differences in SPA development by referring to population aging on a district level. Two different SPA were examined: Associating the own aging process with either ongoing development or with physical losses. Results show that population aging influenced the level of and changes in the SPA ongoing development as well as level in the SPA physical losses, whereas population aging did not predict change in the SPA physical losses. Consistently, individuals living in districts characterized by a fast aging population, high average age, and low proportion of 80+ (C1) reported more negative SPA for both facets in 2008. Change in SPA physical losses was not associated with differences in population aging, but individuals living in districts with slow population aging, low average age, and low proportion of 80+ (C2) showed a more positive development of SPA ongoing development over time as compared to individuals living in regions with average population aging (C4). In line with the findings of North and Fiske (2015), results suggest that persons have more negative SPA if they live in regions where population aging happens faster. A relatively fast change in the composition of a population may affect the perception of how well needs are met in regard to local circumstances. Changes in leisure time facilities, or regarding health care (e.g., because of retiring General Practitioners) and also the mere age-related changes in the social structure of the neighborhood may affect the living conditions of older adults in these areas and pronounce negative aspects and barriers in old age, resulting in more negative perceptions of aging. Investing in infrastructure and in leisure time facilities for older persons may contribute to the perception of having opportunities for ongoing development in old age among the inhabitants of these regions. However, to accomplish this, regions need to have sufficient financial and economic strength. Fast population aging in combination with low economic strength may even amplify negative changes in SPA. On an individual level, it has already been shown that financial resources make a difference for SPA and their consequences (Craciun, Gellert, & Flick, 2017). Vitman, Iecovich, & Alfasi (2014) also demonstrated that social integration of older adults is associated with characteristics of neighborhoods, such as socioeconomic status, ageism, or perceived limitations of outdoor mobility, in Tel-Aviv. There is, however, a need for more studies investigating the interplay of contextual variables and psychological phenomena. Also, positive effects of regional characteristics on SPA were observed in this study. Persons living in regions with slow population aging (e.g., relatively low average age and small proportion of old-old adults) as compared to the German average showed a positive development in their SPA ongoing development. Possibly, these regions are able to better adapt to a slowly growing number of older adults as compared to regions with faster population aging. A lower proportion of older adults, and in particular of old-old adults, may result in fewer negative role models and, therefore, more positive SPA. This would be in line with research showing, that perception of older adults depends on societal roles of older adults: Older persons are perceived as more competent in countries with a higher proportion of older adults participating in paid and volunteer work (Bowen & Skirbekk, 2013). In particular, regions with higher proportions of healthy older adults may benefit from these positive effects. Additionally, social contact and cohesion may be different in those communities where the population structure has been shifted gradually over time. As suggested in the model by Li (2003), intergenerational interactions may shape development of individuals. Future research should investigate which mechanisms might underlie this relationship. Change in SPA physical losses, in contrast, was not affected by population aging as operationalized in the clusters. Löckenhoff and colleagues (2009) argue that context effects are strongest for the aspects of aging perceptions that are least influenced by age-related biological changes, which was supported by the finding that physical perceptions of aging were less related to country level indicators than other facets such as socioemotional or family-related views of aging. Possibly, expectations of physical loss with aging are more likely to be influenced by individual factors such as health behaviors or health events than by contextual circumstances. In sum, our findings add to the previous research by extending what has been shown for age stereotypes in different countries (cf. North & Fiske, 2015): Population aging is associated with changes in self-perceptions of aging. To some extent, results contradict the prediction of the contact hypothesis (cf. Allport, 1954; Pettigrew & Tropp, 2006). However, little is known about mechanisms of population aging on social contact and cohesion. For instance, Löckenhoff et al. (2009) found that a higher proportion of older adults in the population was not related to more frequent intergenerational contact with young adults. Limitations Although population aging can be considered similar in different districts of the same cluster, this does not imply that other environmental variables such as infrastructure or opportunities for leisure and engagement are similar. The district-level is the smallest regional level with available data on population aging in Germany. Nevertheless, living conditions within one district can differ to a considerable extent. In addition, individuals may differ in the degree of use of services provided by neighboring regions, and consequently may be more or less exposed to the population composition. However, this limitation speaks for less strong contextual effects and therefore even underlines the small but significant effects of differences in population aging on SPA found in this study. Generalizations of results to other than German regions should be drawn with caution. In particular, country-specific age stereotypes and country-specific legislations (e.g., welfare state) may drive some of the observed results. In addition, countries may differ in social change of SPA (Germany shows a positive pattern over time; Beyer et al., 2017) which may affect the results. Future studies should compare regions with similar population aging patterns between different countries. The clusters used reflect a share of different population aging processes. It remains unclear, if certain aspects are more strongly related to SPA than others. A high proportion of old-old people may challenge regions differently as compared to a high proportion of young-old people. In addition, other regional aspects, not reflected in the clusters, may be relevant for the SPA of the inhabitants. This study combines the investigation of social and intraindividual change in a sample with a broad age range (40–85 years). Even though FIML counteracts the effects of sample attrition, interpretations regarding intraindividual change in SPA should be made with caution. However, the combination of social and intraindividual change has the advantage of a large sample adequately representing districts and the four clusters in Germany as well as reflecting the effects of population aging over 12 years for both social and intraindividual changes in SPA. Furthermore, there might be time periods in which contextual factors have a stronger effect on individual SPA (e.g., availability of health care institutions after a health event; engagement opportunities after retirement). Continuous, closely-timed SPA assessment may help capture those time periods and allow the identification of potential intervention targets. Although indicators of wealth and health care opportunities were included as covariates, underlying mechanisms, such as leisure opportunities and economic living conditions in the districts were not fully captured in our model and may drive the random effects of the multilevel models. Future studies should additionally try to capture these district characteristics to further explain the district level variance in level and change of SPA. Conclusion As most societies face population aging, it is important to investigate its effects on individual development and aging. The current study focuses on contextual effects on self-perceptions of aging as one important psychological resource for health, well-being and longevity in later life. Results suggest that SPA do not develop independently from population aging on the district level. Rather, the pace of population aging in different regions of one single country can contribute to individual changes in positive and negative SPA. This finding complements previous cross-cultural comparisons. Districts facing fast population aging should address possible negative effects by providing attractive opportunities for families with young children, by increasing efforts to maintain and improve good infrastructure (e.g., public transport, health care, and shopping facilities), and by providing opportunities for intergenerational contact and mutual support. Additionally, health communication and public health messages that emphasize positive aspects of aging (e.g., wisdom, positive older role models) may counteract negative effects of fast population aging. Results imply that regional differences and potential contextual opportunities for aging well should be taken into account when trying to promote positive perceptions of aging. Funding This work was supported by the German Federal Ministry of Family Affairs, Senior Citizens, Women and Youth (grant number 301-6083-05/003*2) and by the German Research Foundation (DFG; grant 632451, granted to Verena Klusmann). Conflict of Interest None reported. Acknowledgments The authors thank Constantin Bolz for his support in data preparation. J. K. Wolff, A.-K. Beyer, S. Wurm, and M. Wiest are members of the Scientific Network “Images of Aging,” funded by the German Research Foundation (DFG; grant 632451, granted to Verena Klusmann). References Allport, G. W. ( 1954). The nature of prejudice . Reading, MA: Addicon Wesley. Beyer, A.-K., Wurm, S., & Wolff, J. K. ( 2017). Älter werden – Gewinn oder Verlust? Individuelle Altersbilder und Altersdiskriminierung. [Growing older – gain or loss? Individual perceptions of aging and age discrimination] In K. Mahne, J. K. Wolff, J. Simonson, & C. Tesch-Römer (Eds.), Altern im Wandel: Zwei Jahrzehnte Deutscher Alterssurvey (DEAS)  (pp. 329– 343). Wiesbaden: Springer VS. Bowen, C. E., & Skirbekk, V. ( 2013). National stereotypes of older people’s competence are related to older adults’ participation in paid and volunteer work. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 68, 974– 983. doi: 10.1093/geronb/gbt101 Google Scholar CrossRef Search ADS   Bronfenbrenner, U. ( 1977). Toward an experimental ecology of human-development. American Psychologist , 32, 513– 531. doi: 10.1037/0003-066x.32.7.513 Google Scholar CrossRef Search ADS   Bronfenbrenner, U. ( 1979). The ecology of human development: Experiments by nature and design . Cambridge, MA: Harvard University Press. Bullinger, M., & Kirchberger, I. ( 1998). Der SF-36 Fragebogen zum Gesundheitszustand. Handbuch für die deutschsprachige Fragebogenversion . Göttingen: Hogrefe Verlag. Bryant, C., Bei, B., Gilson, K. M., Komiti, A., Jackson, H., & Judd, F. ( 2016). Antecedents of attitudes to aging: A study of the roles of personality and well-being. The Gerontologist , 56, 256– 265. doi: 10.1093/geront/gnu041 Google Scholar CrossRef Search ADS PubMed  Craciun, C., Gellert, P., & Flick, U. ( 2017). Aging in precarious circumstances: Do positive views on aging make a difference? The Gerontologist , 57, 517– 528. doi: 10.1093/geront/gnv135 Google Scholar PubMed  Cagny, K. A., & Wen, M. ( 2008). Social capital and aging-related outcomes. In I. Kawachi, S. V. Subramanian, & D. Kim (Eds.), Social capital and health  (pp. 239– 258). New York: Springer. Google Scholar CrossRef Search ADS   Dannefer, D. ( 1992). On the conceptualization of context in developmental discourse: Four meanings of context and their implications. In D. L. Featherman, R. M. Lerner, & M. Perlmutter (Eds.), Life-span development and behavior  (pp. 83– 110). Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers. Donlon, M. M., Ashman, O., & Levy, B. R. ( 2005). Re-vision of older television characters: A stereotype-awareness intervention. Journal of Social Issues , 61, 307– 319. doi: 10.1111/j.1540-4560.2005.00407.x Google Scholar CrossRef Search ADS   Gerstorf, D., & Ram, N. ( 2012). Late life. In S. Krauss Whitebourne, & M. J. Sliwinski (Eds.), Wiley-Blackwell handbook of adulthood and aging  (pp. 49– 71). Oxford, UK: Wiley-Blackwell. Google Scholar CrossRef Search ADS   Giles, H., Noels, K. A., Williams, A., Ota, H., Lim, T. S., Ng, S. H., … Somera, L. ( 2003). Intergenerational communication across cultures: Young people’s perceptions of conversations with family elders, non-family elders and same-age peers. Journal of Cross-Cultural Gerontology , 18, 1– 32. doi: 10.1023/a:1024854211638 Google Scholar CrossRef Search ADS PubMed  Kim, J. E., & Moen, P. ( 2002). Retirement transitions, gender, and psychological well-being: A life-course, ecological model. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 57, P212– P222. doi: 10.1093/geronb/57.3.P212 Google Scholar CrossRef Search ADS   Klaus, D., Engstler, H., Mahne, K., Wolff, J. K., Simonson, J., Wurm, S., & Tesch-Römer, C. ( 2017). Cohort profile: The German Ageing Survey (DEAS). International Journal of Epidemiology . doi: 10.1093/ije/dyw326 Kleinspehn-Ammerlahn, A., Kotter-Grühn, D., & Smith, J. ( 2008). Self-perceptions of aging: Do subjective age and satisfaction with aging change during old age? The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 63, P377– P385. doi: 10.1093/geronb/63.6.P377 Google Scholar CrossRef Search ADS   Kornadt, A. E., & Rothermund, K. ( 2011). Contexts of aging: Assessing evaluative age stereotypes in different life domains. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 66, 547– 556. doi: 10.1093/geronb/gbr036 Google Scholar CrossRef Search ADS   Kornadt, A. E., Voss, P., & Rothermund, K. (2017). Age stereotypes and self-views revisited: Patterns of internalization and projection processes across the life span. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 72, 582–592. doi:10.1093/geronb/gbv099 Kwak, M., Ingersoll-Dayton, B., & Burgard, S. ( 2014). Receipt of care and depressive symptoms in later life: The importance of self-perceptions of aging. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 69, 325– 335. doi: 10.1093/geronb/gbt128 Google Scholar CrossRef Search ADS   Levy, B. R. ( 2008). Rigidity as a predictor of older persons’ aging stereotypes and aging self-perceptions. Social Behavior and Personality , 36, 559– 570. doi: 10.2224/sbp.2008.36.4.559 Google Scholar CrossRef Search ADS   Levy, B. ( 2009). Stereotype embodiment: A psychosocial approach to aging. Current Directions in Psychological Science , 18, 332– 336. doi: 10.1111/j.1467-8721.2009.01662.x Google Scholar CrossRef Search ADS PubMed  Li, S. C. ( 2003). Biocultural orchestration of developmental plasticity across levels: The interplay of biology and culture in shaping the mind and behavior across the life span. Psychological Bulletin , 129, 171– 194. doi: 10.1037/0033-2909.129.2.171 Google Scholar CrossRef Search ADS PubMed  Löckenhoff, C. E., De Fruyt, F., Terracciano, A., McCrae, R. R., De Bolle, M., Costa, P. T. Jr,… Yik, M. ( 2009). Perceptions of aging across 26 cultures and their culture-level associates. Psychology and Aging , 24, 941– 954. doi: 10.1037/a0016901 Google Scholar CrossRef Search ADS PubMed  Macia, E., Lahmam, A., Baali, A., Boëtsch, G., & Chapuis-Lucciani, N. ( 2009). Perception of age stereotypes and self-perception of aging: A comparison of French and Moroccan populations. Journal of Cross-Cultural Gerontology , 24, 391– 410. doi: 10.1007/s10823-009-9103-0 Google Scholar CrossRef Search ADS PubMed  McCann, R. M., & Keaton, S. A. ( 2013). A cross cultural investigation of age stereotypes and communication perceptions of older and younger workers in the USA and Thailand. Educational Gerontology , 39, 326– 341. doi: 10.1080/03601277.2012.700822 Google Scholar CrossRef Search ADS   Menning, S., Nowossadeck, E., & Maretzke, S. ( 2010). Regionale Aspekte der demografischen Alterung. [Regional aspects of demographic aging]. Report Altersdaten , 1–2, 1– 56. Retrieved from http://nbn-resolving.de/urn:nbn:de:0168-ssoar-369378. Moor, C., Zimprich, D., Schmitt, M., & Kliegel, M. ( 2006). Personality, aging self-perceptions, and subjective health: A mediation model. International Journal of Aging & Human Development , 63, 241– 257. doi: 10.2190/AKRY-UM4K-PB1V-PBHF Google Scholar CrossRef Search ADS PubMed  Mun, E. Y., von Eye, A., & White, H. R. ( 2009). An SEM approach for the evaluation of intervention effects using pre-post-post designs. Structural Equation Modeling: A Multidisciplinary Journal , 16, 315– 337. doi: 10.1080/10705510902751358 Google Scholar CrossRef Search ADS PubMed  North, M. S., & Fiske, S. T. ( 2015). Modern attitudes toward older adults in the aging world: A cross-cultural meta-analysis. Psychological Bulletin , 141, 993– 1021. doi: 10.1037/a0039469 Google Scholar CrossRef Search ADS PubMed  Pettigrew, T. F., & Tropp, L. R. ( 2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology , 90, 751– 783. doi: 10.1037/0022-3514.90.5.751 Google Scholar CrossRef Search ADS PubMed  Putnam, R. D. ( 2000). Bowling alone . New York: Simon & Schuster. Rentfrow, P. J., & Jokela, M. ( 2016). Geographical psychology: The spatial organization of psychological phenomena. Current Directions in Psychological Science , 25, 393– 398. doi: 10.1177/0963721416658446 Google Scholar CrossRef Search ADS   Sargent-Cox, K. A., Anstey, K. J., & Luszcz, M. A. ( 2012). Change in health and self-perceptions of aging over 16 years: The role of psychological resources. Health Psychology: Official Journal of the Division of Health Psychology, American Psychological Association , 31, 423– 432. doi: 10.1037/a0027464 Google Scholar CrossRef Search ADS PubMed  Schafer, J. L., & Graham, J. W. ( 2002). Missing data: Our view of the state of the art. Psychological Methods , 7, 147– 177. doi: 10.1037/1082-989x.7.2.147 Google Scholar CrossRef Search ADS PubMed  Silverstein, M., Parrott, T. M., Angelelli, J. J., & Cook, F. L. ( 2000). Solidarity and tension between age groups in the United States: Challenge for an aging America in the 21st century. International Journal of Social Welfare , 9, 270– 284. doi: 10.1111/1468–2397.00139 Google Scholar CrossRef Search ADS   Steverink, N., Westerhof, G. J., Bode, C., & Dittmann-Kohli, F. ( 2001). The personal experience of aging, individual resources, and subjective well-being. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 56, P364– P373. doi: 10.1093/geronb/56.6.P364 Google Scholar CrossRef Search ADS   Thompson, E. H. Jr, & Weaver, A. J. ( 2016). Making connections: The legacy of an intergenerational program. The Gerontologist , 56, 909– 918. doi: 10.1093/geront/gnv064 Google Scholar CrossRef Search ADS PubMed  UNESCO.( 2012). International Standard Classification of Education ISCED 2011 . Montreal: UNESCO Institute for Statistics. Vitman, A., Iecovich, E., & Alfasi, N. ( 2014). Ageism and social integration of older adults in their neighborhoods in Israel. The Gerontologist , 54, 177– 189. doi: 10.1093/geront/gnt008 Google Scholar CrossRef Search ADS PubMed  Westerhof, G. J., Miche, M., Brothers, A. F., Barrett, A. E., Diehl, M., Montepare, J. M., … Wurm, S. ( 2014). The influence of subjective aging on health and longevity: A meta-analysis of longitudinal data. Psychology and Aging , 29, 793– 802. doi: 10.1037/a0038016 Google Scholar CrossRef Search ADS PubMed  Wurm, S., Tesch-Römer, C., & Tomasik, M. J. ( 2007). Longitudinal findings on aging-related cognitions, control beliefs, and health in later life. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 62, P156– P164. doi: 10.1093/geronb/62.3.p156 Google Scholar CrossRef Search ADS   Wurm, S., Wolff, J. K., & Schüz, B. ( 2014). Primary care supply moderates the impact of diseases on self-perceptions of aging. Psychology and Aging , 29, 351– 358. doi: 10.1037/a0036248 Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Journal

The GerontologistOxford University Press

Published: Feb 1, 2018

There are no references for this article.