Preparation for old age in different life domains

Abstract

We investigated preparation for age-related changes from a multidimensional, life span perspective and administered a newly developed questionnaire to a large sample aged 30–80 years. Preparing for age-related changes was organized by life domains, with domain-specific types of preparation addressing obstacles and opportunities in the respective domains. Preparing for a third (focusing on activities, leisure, work, fitness, appearance) and a fourth age (focusing on emergencies, dependence/independence, housing, financial arrangements) emerged as superordinate categories of preparation. Different age gradients were obtained for the factors, the former peaking around the age of 65, whereas the latter increased linearly up to the age of 80. Furthermore, preparation factors were characterized by distinct personality profiles. The findings attest to the importance of a differentiated view on preparation for age-related changes and its relevance across the life span.

Keywords

Aging is a major challenge in the life span: As people grow older and move from youth through middle into older age, things in our lives change and one has to face new challenges. This might start with small things such as discovering bodily changes like wrinkles, include role transitions like leaving the workplace and retiring, and might go as far as accepting that one cannot live an independent life anymore and from a certain point on has to rely on help from others. To adapt to these age-related changes and master these challenges, one has to mobilize and re-distribute resources such as money, social support or simply effort to assure that one can still live the life one wants to live—at least to a certain degree. Especially in cultures where societal and familial systems do not provide an inbuilt safety net for their aging populations, whether one is prepared to meet the challenges of aging or not, remains mostly one’s own responsibility (Denton et al., 2004; Ekerdt, 2004; Hershey, Henkens, & Van Dalen, 2007).

Jacobs-Lawson, Hershey and Newkam (2004) define planning for old age as “an effort to ensure major problems will not develop at some point in the distant future” (p. 57), and even though it is still not entirely clear which kinds of preparation are beneficial under what circumstances (Donaldson, Earl, & Muratore, 2010), there is a multitude of studies that have demonstrated the positive effects of preparation efforts: People who prepare for age-related changes are less anxious about retirement (MacEwen, Barling, Kelloway, Higginbottom, 1995), report less depression after a 2-year interval (Sörensen, Mak, Chapman, Duberstein, & Lyness, 2012), and more well-being in old age (Noone, Stephens & Alpass, 2009; Pinquart & Sörensen, 2002), are actually better prepared (Noone, Stephens & Alpass, 2010) and healthier (Topa, Moriano, Depolo, Alcover, & Morales, 2009) than people who plan less. Adams and Rau (2011) also highlight the positive aspects of (retirement) preparation, but note, as well, that there are still many people that refrain from preparing for their future in old age (cf. Ekerdt, Hackney, Kosloski, & DeViney, 2001; Jacobs-Lawson et al., 2004). Therefore, understanding when and how people plan for older age (or not) is a crucial question to understand and contribute to a positive aging experience.

The multifaceted nature of preparation for old age

There is already a large amount of research concerned with preparation for old age (for a recent review, see Adams & Rau, 2011, and for a comprehensive sociological approach, see Street & Desai, 2011). However, up to today, some interesting and relevant questions have remained unanswered. Most researchers frame their studies as “preparation for retirement,” and therefore, they almost exclusively investigate preparation for events and challenges that are directly or indirectly linked with the exit from the workplace (Muratore & Earl, 2010; Noone et al., 2009, 2010). However, even though the exit from the workplace may be one of the most salient transitions one has to face in the second half of life, affecting many life domains and requiring intensive preparation, this conceptualization might be too narrow to comprehensively capture preparation for old age over the life span (Denton et al., 2004). One reason for that is that for a large group of people, there is no fixed retirement transition at a certain, pre-defined age. If one thinks of housewives, self- or unemployed persons, planning for “retirement” defined as leaving the paid workforce, does not apply to them, but they still have to prepare for other age-related changes. Relatedly, even the typical employee may be concerned about other changes or transitions that are expected to occur in old age but are unrelated to retirement (e.g. care, loneliness, sickness, grandchildren, functional decline, reduced mobility, weariness, time management, leisure). What can be seen by these examples is that preparation for age-related changes is a multidimensional construct that includes more than just the financial domain, which has been the focus of many studies so far (Jacobs-Lawson et al., 2004). A notable exception of this strong focus on financial and retirement-related preparation research is concerned with preparation for future care needs (cf. Pinquart & Sörensen, 2002). The studies conducted in this realm offer a detailed and in-depth investigation, reporting antecedents and consequences of planning in this domain, and also instruments to assess preparation for care needs (Sörensen & Pinquart, 2001). However, we think that a multidimensional, life course perspective that includes preparation in a diverse set of life domains, may offer a more comprehensive perspective (Ekerdt, 1998; Street & Desai, 2011; Wang & Shultz, 2010).

There already are some studies that have looked at preparation for age-related changes from a multidimensional perspective. Some instruments were developed that also included items regarding the domains of health, leisure, lifestyle and work (Muratore & Earl, 2010; Noone et al., 2010; Petkoska & Earl, 2009). Nevertheless, these instruments still had a strong focus on financial preparation (two thirds of the items were related to financial preparations) and/or were administered only to people who were still employed in the workforce or were approaching retirement. Up to now, the most comprehensive investigation of preparation in different life domains comes from Hershey, Brown, Jacobs-Lawson and Jackson (2001), who conducted a qualitative study that asked participants about their perceptions of important retirement-planning decisions. They identified a framework of seven dimensions or life domains (health issues; financial issues; work, career and employment; housing issues; estate planning; leisure & recreation; social & interpersonal), but no instrument was developed on the basis of this taxonomy that would allow for a quantitative assessment of these different facets of preparation.

Preparation for age-related changes as a life-long process

A multidimensional conceptualization of preparation as “preparation for age-related changes” rather than “retirement preparation” seems to be the more comprehensive approach if one considers that developmental change occurs and is perceived throughout the entire life span (Baltes, 1997; Heckhausen, Dixon, & Baltes, 1989). Even though preparing for old age increases with age (Ekerdt, Kosloski, & DeViney, 2000; Jacobs-Lawson et al., 2004), the foundations can be laid already in younger and middle ages (for an example in the financial domain, see Hira, Rock, & Loibl, 2009). In a study by Hershey et al. (2001), participants were asked about the age at which they would recommend to start planning. The mean age that was given was 36.1 years, and asked what they would have changed in their own planning behaviour if they had the chance, many of the respondents indicated that they would have started planning earlier.

Furthermore, if one also considers that the post-retirement phase might extend over a period of 20 years for some people (i.e., in Germany, the mean retirement age is about 64), preparation does not stop with retirement but continues up to the very high ages (Donaldson et al., 2010). The study of preparation for age-related changes over the adult life span therefore might include preparations for an active “third age,” where the loss of the work role and its substitution with leisure activities or bridge employment are relevant, as well as those for a “fourth age” in which the deterioration of one’s health, dependence/independence, generativity issues, and closeness to death might become more important. Preparations in later stages of the lifespan might focus on desired (or undesired) changes in living arrangements like moving closer to one’s children or to an assisted facility so that one can still live a decent life in the face of experiencing age-related constraints, and also on times in which one is no longer able to make decisions on one’s own such as for future care needs in case of dementia or other severe illness (Baltes & Smith, 2003; Lang, Baltes, & Wagner, 2007; Sörensen et al., 2012).

A differential engagement in different life domains across the life span has been demonstrated before with the concept of personal life investment (PLI; Schindler & Staudinger, 2008; Schindler, Staudinger, & Nesselroade, 2006). In their research, the authors assessed current motivational engagement in various life domains, and came up with a distinction between obligatory life investment (e.g. reflected in the domains health, independence, and life reflection) as opposed to optional life investment (e.g. leisure, occupation, sexuality). Whereas optional life investment declined with age, obligatory investment remained rather stable up to the very high ages (Schindler et al., 2006). Building on these results, we wanted to investigate whether such a distinction between different phases of older age is not only visible in current motivational engagement, but also in preparing for anticipated future age-related changes in different life domains.

Thus, the guiding idea of our study was that preparation for age-related changes is organized by life domains as well. In addition to research on PLI, recent research on perceptions of old age and aging has demonstrated that personal as well as stereotypical views on aging are context-dependent and domain-specific (e.g. Casper, Rothermund, & Wentura, 2011; Kornadt & Rothermund, 2011a). Expectations of what will change in old age and how people will change when they get older are not global and abstract, but rather embedded into the particular concrete life circumstances in different life domains. Accordingly, preparation for age-related changes should be concerned with specific hopes and fears that exist with regard to possible changes, opportunities, and dangers that might occur in particular life domains (e.g. the fear of losing one’s spouse [partnership], the possibility to live in a place that is close to one’s children and grandchildren [family], the prospect to have more spare time after retirement [leisure, engagement], anticipated deterioration of mental and bodily functioning at the end of life [health, autonomy], the necessity to determine what should happen in case of death or disability [emergencies]). A first research question thus was to investigate the dimensional structure of preparation for age-related changes, and to test whether preparation intentions and activities are systematically organized by life domains.

In addition, we wanted to investigate qualitative and quantitative differences in preparation across the adult life span. As intimated above, we assume that preparation for age-related changes is not restricted to old age itself, but should be a concern throughout life: Middle-aged and even younger adults should also engage in preparation for changes that are expected to occur during later life. However, with decreasing distance to the onset of the respective phase of life, preparation should become more urgent and thus increase. Theories of motivation and action have proposed that the incentive value and the resulting motivational force of a positive or negative outcome on behaviour is an inverse function of the perceived time distance at which the respective outcome is supposed to occur (Green & Myerson, 2004). Distant events thus have less impact on behaviour than outcomes that are expected to occur in the near future. Based on this temporal discounting framework, we expect a complex pattern of age-related changes for different dimensions of preparation activities: Depending on the nature of change and on the age range at which changes are expected to occur in a life domain, preparation should peak at different points in life. Specifically, changes related to retirement (entering into third age) should affect certain life domains more than others and should lead to high levels of preparation intentions fairly early in life, whereas age-related changes expected at the end of life (entering into fourth age) should affect other domains for which preparation should continue to increase in intensity until very old age.

Overview and aims of the study

In the present study, we wanted to explore preparation for old-age-related changes from a multidimensional perspective, covering a large part of the adult life span. A major aim of our study was thus to develop a new measure of preparation that allowed us to assess the full range of preparation activities and to investigate the organizing principles according to which preparation is represented and regulated across the life span. We assessed levels of preparation intentions for age-related changes with respect to nine different life domains in a sample ranging from the age of 30 to 80 years. This allowed us to investigate the dimensional structure of preparation and to compare age gradients for the different forms of preparing for age-related changes. Since preparation for old age, for future care needs, and PLI have been related to personality (Prenda & Lachman, 2001; Schindler & Staudinger, 2008; Sörensen, Duberstein, Chapmann, Lyness, & Pinquart, 2008), we also assessed personality variables in order to investigate whether different types of preparation reflect corresponding personality profiles.

Method

Sample

The sample consisted of N = 951 persons (51% female) aged 30–80 years (M = 54.8 years, SD = 14.6). A large part of the sample (n = 769) was recruited via random sampling through local registry offices in two middle-sized cities in Germany (Jena and Erlangen). In order to generate a diverse sample that was balanced according age group and gender, the sample was disproportionately stratified for birth cohort (years of birth 1929–1938, 1939–1948, 1949–1958, 1959–1968, 1969–1978; age at time of measurement: 30–80 years), gender, and also place of residence.

Participants were approached with a letter, asking them whether they wanted to participate in the study. They could consent by sending back an enclosed postcard and were then sent the questionnaires via mail. They filled them in at home and returned them, again via mail. The remaining participants (n = 182, age range 30–80 years) were recruited via convenience and snowball sampling. Participants were approached by a research assistant, and upon consenting to participation, were mailed or handed out the questionnaire and returned it to the research assistant personally, or again, per mail. The questionnaire for this subsample additionally included a measure of personality, which was not part of the questionnaire that was sent to the larger subsample.

Sample characteristics for the total sample are displayed in Table 1. Comparing our sample with data from the German population in this age range (German Federal Statistics Office), we found that our sample was more highly-educated (probably due to the fact that we sampled in two University cities), and more likely to be married. There were no pronounced differences with regard to income.

Table 1.

Sample description.

	total sample	aged 30–40	aged 41–50	aged 51–60	aged 61–70	aged 71–80
	N = 951	n = 193	n = 215	n = 180	n = 165	n = 198
Sex N female (%)	488 (51)	97 (50)	103 (47)	100 (56)	84 (51)	104 (53)
Age M (SD)	54.83 (14.58)	34.90 (3.10)	45.51 (2.91)	55.34 (2.81)	65.68 (2.94)	74.85 (2.69)
Marital status N (%)
single	94 (10)	44 (23)	27 (13)	13 (7)	3 (2)	7 (4)
married or partnership	695 (73)	141 (73)	147 (68)	144 (80)	132 (80)	131 (66)
separated or divorced	95 (10)	5 (3)	35 (16)	14 (8)	20 (12)	20 (10)
widowed	56 (6)	0 (0)	0 (0)	8 (4)	9 (5)	39 (20)
other or n/a	12 (1)	3 (2)	7 (3)	8 (4)	2 (1)	1 (1)
Education N (%)
Apprenticeship or vocational school	317 (33)	66 (34)	79 (37)	66 (37)	47 (28)	59 (3)
School for master craftspeople	170 (18)	20 (10)	38 (18)	29 (16)	36 (22)	47 (24)
University degree	401 (42)	95 (49)	80 (37)	82 (46)	77 (47)	68 (34)
Other or n/a	62 (7)	12 (6)	18 (8)	3 (2)	5 (3)	24 (12)
Monthly net household income (€) M (SD)	2643.74 (1429.79)	2679.29 (1319.21)	3037.08 (1784.38)	2996.23 (1482.34)	2399.76 (1228.86)	2072.57 (915.89)

Measures

Preparation for age-related changes

We developed a new measure to assess preparation for age-related changes in different life domains. To make sure we included a relevant set of life domains, we first conducted an interview study with older persons (N = 14) to assess in which life domains age-related changes are expected and experienced; details on this study are described in Kornadt and Rothermund (2011a). Furthermore, we adapted some of the domains that were previously used in the literature (Hershey et al., 2001) and added domains we deemed relevant but which were not mentioned or used before. This process yielded nine different life domains: Finances (FIN), Emergencies and Exceptional Circumstances (EM), Mental and Physical Fitness (FIT), Housing (HOUS), Looks and Appearance (LOOK), Social relationships (SOC), Health (HEALTH), Leisure Activities and Lifestyle (LEIS), Work and Employment (WORK).

For each domain, we formulated four general statements to assess different aspects of preparation for age-related changes: active preparation (“I actively prepare for age-related changes in this domain”), thoughts and preoccupation with the topic (“I think about and am preoccupied with preparation in this domain”), gathering and exchanging information (“I gather information and communicate with others about preparation in this domain”), and worry (“I am worried about aging in this domain”) resulting in 36 items altogether (nine domains × four items). Participants gave their responses to each item on a 4-point scale with the options “not at all,” “a little,” “quite a bit” and “a lot.” We decided to assess preparation with different items in order to account for the fact that preparation not only encompasses actually doing something, but also gathering knowledge and making up one’s mind about the topic (e.g. Ekerdt et al., 2001; Jacobs-Lawson et al., 2004; Mak & Sörensen, 2012; Pinquart & Sörensen, 2000; Stawski, Hershey, & Jacobs-Lawson, 2007), and also to obtain a more reliable scale. The active preparation item was illustrated with several concrete examples for each domain in order to provide a more specific understanding of how preparation could look like in the respective domain (see Table 2). We refrained from asking participants to indicate the absolute amount of preparation separately for each of the specific activities because we were interested in the general level of preparation, aggregating across different forms of behavioural and mental activities that are conducted with the intention to be prepared for age-related changes that might occur within each domain.

Table 2.

Life domains and examples provided for the assessment of preparation intentions in each domain.

Life domain	Example
FIN	I actively prepare for my financial situation in old age (i.e., government-provided saving opportunities, pension funds, saving money).
EM	I actively prepare for emergencies or exceptional situations in old age (i.e., will, health care proxy, advance health care directive).
FIT	I actively prepare to maintain my mental and physical fitness in old age (i.e., physical and mental exercise).
HOUS	I actively prepare for my housing situation in old age (i.e., age-appropriate remodelling, buying real estate, agreements with children/partner/friends).
LOOK	I actively prepare for my appearance in old age (i.e., beauty products, clothing, diet, auxiliary devices).
SOC	I actively prepare for my social relationships in old age (i.e., networking, integration in social groups).
HEALTH	I actively prepare to maintain my health in old age (i.e., medical check-ups, avoidance of health-damaging behaviour).
LEIS	I actively prepare for activities in old age (i.e., hobbies, looking for new tasks and recreational activities).
WORK	I actively prepare for a successful career in old age (i.e., further qualification, prevention of work-related illness and disability).

Note. FIN = Finances, EM = Emergencies and exceptional circumstances, FIT = Mental and physical fitness, HOUS = Housing, LOOK = Looks and appearance, SOC = Social relationships, HEALTH = Health, LEIS = Leisure activities and lifestyle, WORK = Work and employment.

Personality

Inter-individual differences regarding the five basic dimensions of personality were assessed with the BFI-K, a 15-item measure that has been demonstrated to reliably measure the BIG-5 in different age groups (Lang, John, Lüdtke, Schupp, & Wagner, 2011).

Data treatment

In order to maintain the full sample of participants for all analyses, single missing values (less than 5% per variable) for the preparation intentions were supplemented by data imputation with the expectation maximization method (Dempster, Laird, & Rubin, 1977).

Results

The structure of preparation for age-related changes

To assess whether preparation for age-related changes is a multidimensional construct, we conducted a principal axis factor analysis with Oblimin rotation on the 36 items designed to measure levels of preparation in different life domains. The KMO criterion was .89, indicating a very good sampling adequacy, and Bartlett’s test of sphericity was significant, χ²(630) = 22,177.60, p < .001, which is an indicator of sufficient inter-correlations among variables. A 10-factor solution explained 74.25% of the variance in the original items (all Eigenvalues ≥ .96),¹ and provided a perfect assignment of the preparation items to the specific domains that corresponded to our a priori grouping; communalities and factor loadings of the 10-factor solution are presented in Table 3. The items that were designed to measure preparation in the different domains grouped together on nine single factors uniquely representing the respective life domain, without substantial cross-loadings. An additional 10th factor showed substantial loadings only for the worry items of the nine domains, which were grouped together on a separate factor independently of life domain content. Levels of worry did not vary systematically between age groups (F[4, 946] = 1.10, p = .36) indicating that the amount of worry across domains is independent of active preparation and planning for age-related changes, and might rather reflect a superordinate facet of personality that is related to neuroticism and anxiety (Rammstedt, 2007), which is why we decided not to focus on the worry factor in the subsequent analyses of preparation intentions.²

Table 3.

Factor loadings, Communalities, and Eigenvalues for the first order factor analysis.

Rotated factor loadings
Item	LEIS	WORRY	WORK	FIN	LOOK	HEALTH	HOUS	EM	FIT	SOC	Communalities
FIN preparation				.51					.44		.61
FIN thoughts				.84							.80
FIN information				.75							.74
FIN worry		.47		.37							.56
EM preparation								.88			.77
EM thoughts								.87			.81
EM information								.84			.81
EM worry		.70									.61
FIT preparation									.69		.69
FIT thoughts									.40		.61
FIT information	.34					.33			.15		.63
FIT worry		.79									.68
HOUS preparation							.80				.75
HOUS thoughts							.80				.82
HOUS information							.74				.81
HOUS worry		.75									.66
LOOK preparation					.85						.77
LOOK thoughts					.89						.85
LOOK information					.81						.83
LOOK worry		.51			.53						.65
SOC preparation										.89	.79
SOC thoughts										.84	.82
SOC information										.75	.81
SOC worry		.81									.68
HEALTH preparation						.73					.71
HEALTH thoughts						.77					.79
HEALTH information						.72					.79
HEALTH worry		.76									.69
LEIS preparation	.73										.71
LEIS thoughts	.85										.85
LEIS information	.76										.83
LEIS worry		.69									.67
WORK preparation			.85								.77
WORK thoughts			.91								.88
WORK information			.84								.84
WORK worry		.47	.59								.66
Eigenvalue	10.25	4.18	2.53	2.09	1.65	1.50	1.30	1.25	1.03	0.96
% Variance	28.47	11.62	7.02	5.79	4.57	4.15	3.61	3.46	2.87	2.68

Note. N = 951. Cross loadings <.30 are suppressed. FIN = Finances, EM = Emergencies and exceptional circumstances, FIT = Mental and physical fitness, HOUS = Housing, LOOK = Looks and appearance, SOC = Social relationships, HEALTH = Health, LEIS = Leisure activities and lifestyle, WORK = Work and employment.

Even though the factor analysis yielded separate factors representing levels of preparation for the different domains, inter-correlations between these factors were still substantial. Therefore, we subjected the regression factor scores for the indicators of preparation levels in the nine domains to a second order principal axis factor analysis with Oblimin rotation to see whether superordinate factors could be identified according to which the life domains can be grouped together.³ Again, sampling adequacy was very good (KMO = .79) and Bartlett’s test was significant χ²(36) = 1,015.71, p < .001. Two factors with an Eigenvalue > 1 were extracted, explaining 41.21% of the variance of the nine first order factors; factor loadings are presented in Table 4. On the first factor, the domains LEIS, WORK, LOOK, SOC, and FIT had substantial loadings, on the second factor, the domains FIN, HEALTH, HOUS, and EM loaded most strongly.

Table 4.

Factor loadings and Eigenvalues for the second-order factor analysis.

	Second-order factors
First-order factors	third age	fourth age
LEIS	.59
WORK	.65
LOOK	.50
FIT	.49
SOC	.56	.33
HEALTH		.47
HOUS		.52
EM		.77
FIN		.71
Eigenvalue	1.07	2.63
% Variance	11.89	29.32

The results of the second-order factor analysis suggest that preparation for age-related changes consists of two separate superordinate components that address domains which are relevant for an active phase of aging, with the main concerns being the renegotiation of roles, providing the conditions for having fulfilling activities and social relationships after leaving ones’ primary employment, and defying age-related changes in appearance. A second, independent, type of preparation addresses considerations in which care, living arrangements, health deterioration and closeness to death play a major role. For the sake of brevity, and based on the premise that these concerns aim at preparing for age-related changes that occur during early vs. later phases of old age, we refer to these second-order factors with the labels of preparation for a “third” and “fourth age” (e.g. Baltes & Smith, 2003; Laslett, 1989). We address the question of how to interpret these second order factors in more detail in the discussion.

Age differences in preparation, and relations to personality

Age differences

To investigate whether differences in the quality and quantity of preparation for old age and aging indeed change as a function of participant age, and thus validate our assumptions about the factors representing “third” and “fourth” age preparations, we subjected the factor scores of the second-order factor analysis to a mixed-model ANOVA with type of preparation (third- vs. fourth-age preparation) as within-, and participant age group (aged 30–40, 41–50, 51–60, 61–70, and 71–80 years), as between-subjects factor. Mean levels of third-age and fourth-age preparation in the different age cohorts are presented in Figure 1. The ANOVA yielded a significant main effect for birth cohort F(4, 946) = 38.27, p < .001, η _p ² = .14. This effect was due to significant linear and quadratic age trends, indicating that both types of preparation and planning were higher for the older participants (t _lin[946] = 12.03, p < .001), but that the slope of the curve was somewhat less steep for the oldest participants, indicating that the general increase in preparation reached its limits in the oldest age groups (t _quad[946] = −3.26, p < .05). The main effect of age group was qualified by a significant interaction with Type of Preparation F(4, 946) = 11.38, p < .001, η _p ² = .05. Dividing the cohort factor into polynomial contrasts revealed that the interaction effect was based on a significant interaction of Type of Preparation with the linear age trend (t _lin[946] = 6.58, p < .001), all other interactions of Type of Preparation with the higher order trends of the cohort factor were not significant, all t < 1.15. Planned contrasts revealed that even though the linear trend was significant for both types of preparation (third age: t _lin[946] = 4.94, p < .001; fourth age: t _lin[946] = 14.08, p < .001), it was much more pronounced for fourth-age preparation. Third-age preparation clearly dominated fourth-age preparation for the younger age groups, whereas fourth-age preparation dominated over third-age preparation in the older age groups (see Figure 1).

Figure 1.

N = 951. Mean values (and standard errors) for the regression factor scores for third and fourth age preparation. * p < .05.

Relation between preparation and personality

To elucidate the basic motivations that underlie the different types of preparation, we correlated the factors with the Big Five personality scales. Correlation coefficients are displayed in Table 5. Conscientiousness predicted both third- and fourth-age preparation, indicating that conscientious persons, who have internalized high quality standards for evaluating their own behaviour and performance, are also motivated to engage in more advanced planning and preparation for age-related changes in all domains. For the extraversion and openness scales, significant positive correlations were obtained for third-age but not for fourth-age preparation, indicating that the motivation to make positive and new experiences drives planning and preparation for an active, third age, but is unrelated to preparations regarding later phases of old age. No significant correlations were found between the neuroticism and agreeableness scales regarding any of the two superordinate preparation factors.

Table 5.

Correlation coefficients of the second-order preparation intentions factors with the Big Five.

	Second-order factors
	third age	fourth age
Big Five
N	−.00	.04
E	.25	.06
C	.20	.19
A	.08	.09
O	.32	.05

Note. N = 179. Numbers in bold font are significant at p < .05. N = Neuroticism, E = Extraversion, C = Conscientiousness, A = Agreeableness, O = Openness.

Discussion

We investigated preparation for age-related changes in different life domains and from a life span perspective, and for that purpose designed a new instrument. Previous research already demonstrated that preparation for age-related changes includes more than just financial aspects (Adams & Rau, 2011; Denton et al., 2004; Muratore & Earl, 2010; Noone et al., 2010; Sörensen & Pinquart, 2001). However, our results go beyond previous studies by taking a large number of life domains into account and by not restricting our questions to the exit from the workplace as the focal point around which preparation activities are centred. Asking people at a relatively young age of adulthood up to 80 years about their preparations for age-related changes in a wide array of life domains yielded distinguishable factors of preparation for each life domain. In line with our expectations, life domains were established as organizing factors of preparation for age-related changes. Simple distinctions between high vs. low levels of preparation for old age are not suitable to cover the complex patterns of planning and engaging in independent and separate forms of preparation for age-related changes. Apparently, preparation activities are driven by specific hopes and fears that are intimately tied to personal conceptions of the specific and changing contexts of living in old age in various life domains. Similar to what has been reported in the recent literature on views on aging, which have also been shown to reflect context-dependent, domain-specific expectations rather than global age-related attributions (Casper et al., 2011; Gluth, Ebner, & Schmiedek, 2010; Kornadt & Rothermund, 2011a), preparation activities are also organized in terms of the life domains to which they refer. Our findings establish distinguishable patterns of preparation for age-related changes that are expected to occur in different domains. These results yield further evidence for the assumption that preparation for age-related changes is a multidimensional construct relevant for people of different ages throughout the adult life span.

Global dimensions of preparation for old age

Two superordinate dimensions of preparation emerged as second-order factors; the first being mainly characterized by activities and planning in domains like leisure, work, appearance, social relationships and fitness, whereas the second focused on those domains that are expected to be more relevant especially in later phases of life, related to issues of independence/dependence, being close to family members, being taken care of in case of emergencies and age-related illness, but also to generativity (housing, emergency situations, finances and health). We thus described the factors in terms of preparation for an active third age (Laslett, 1989), and for a fourth age in which one has to cope with restrictions and losses that might occur at the end of life (Baltes & Smith, 2003). Apparently, life domains can be grouped according to their association to changes that are expected to occur earlier in old age and that mostly reflect the problems and opportunities that are related to the transition into retirement, or whether they revolve around issues that are related to issues of support, dependency, decline, and decent forms of living in the face of age-related constraints that could characterize terminal phases of one’s life.

The validity of our distinction is substantiated by additional findings highlighting differential relations of third- and fourth-age preparation to other variables. A first interesting result comes from examining the mean values of preparation for people of different age groups: Levels of preparation for third age were lowest among the youngest participants and increased up to the age of 60–70 years, but a reversal of this trend was observed for the oldest age group (> 70 years). Although a cross-sectional observation, this inversely u-shaped age gradient for third-age preparation closely resembles age-related changes in levels of active compensatory efforts aiming at the maintenance of psychological functioning, and at the prevention of losses that also show a similar pattern of increase up to the retirement age followed by a decline in old age (Rothermund & Brandtstädter, 2003b; see also Freund & Baltes, 1998; Schindler et al., 2006). This finding might indicate that active preparation depending on the investment of effort and time might reach its limits in old or very old age (Brandtstädter & Rothermund, 2003). Fourth-age preparations however, were highest for the oldest participants. This finding indicates that preparation for age-related changes and events (including care and support from others as well as death and dying) is a concern even for the very old. It also highlights the fact that preparing for age-related changes might also comprise forms of preparation that provide a basis for accepting, rather than preventing, age-related changes. These forms of preparation are conceptually related to wisdom and accommodative forms of coping (Baltes & Freund, 2003; Baltes, Glück, & Kunzmann, 2002; Brandtstädter & Rothermund, 2002; Brandtstädter, Rothermund, Kranz, & Kühn, 2010; Tornstam, 1997).

The question remains whether our participants also perceived this distinction in terms of two phases of old age, since one could assume that the notion of “third” and “fourth” age represent scientific discourse that might not be widely shared among laypersons. We think that even though it might not be the case that people perceive old age to be “divided” into two distinct phases, the notion of an active age, that is more a continuation of adult life, and a more dependent old age, which marks the actual transition to old age and senescence is widely transported in the media (van Dyk, Lessenich, Denninger, & Richter, 2013) and is also to a certain degree represented in people’s subjective conceptions (Graefe, van Dyk, & Lessenich, 2011). Relatedly, previous findings revealed that life domains differ systematically with regard to the time point at which most age-related changes are assumed to occur (Kornadt & Rothermund, 2012). In particular, most of the very late (> 80 years) changes are expected to occur in the health domain, probably comprising functional losses, emergencies and issues of dependence/independence, whereas age-related changes in other domains (leisure activities, work, relations to friends and acquaintances) are reported mostly at an earlier time point (60–70 years), indicating some relation to a more active phase of old age. So, even though we cannot claim that participants explicitly share the idea of a “third” and “fourth” age that are consecutive in nature and unavoidable in one’s life span, some kind of distinction between goals, expectations and hopes or fears that are tied to more active and more dependent phases of one’s life seem to be on people’s minds. We thus feel positive that naming our second order factors as representing third- and fourth-age preparation captures an important aspect of what people actually perceive when thinking about later phases of life.

However, we should highlight that the labelling of the second-order factors in terms of preparing for a third vs. fourth age reflects only a post hoc interpretation of what preparation in the two groups of life domains might refer to and that there are alternative explanations as to what these factors actually represent. Drawing on the assumption of optional and obligatory life investment, the second-order factors can also be interpreted according to this distinction. Another possibility might be that some of the domains are more closely related to hopes and opportunities (e.g. leisure, social relationships), whereas other domains are more closely related to threats and fears (e.g. emergencies, health). However, we do not think that the life domains that were mapped onto the two second-order factors can be unambiguously interpreted as one or the other. Most of the domains comprise optional and obligatory aspects of preparation, as well as hopes and fears that might be relevant when people think about how to prepare for life in old age. In order to more clearly identify what these two forms of preparation actually represent, more research is needed that takes specific correlates, determinants, and outcomes into account.

Age vs. cohort effects

The different age trajectories that were found for preparation regarding third and fourth age support our general predictions that (a) preparation for age-related changes reflects a process of developmental regulation that starts much earlier than old age itself, and that (b) varies in intensity as a function of the perceived distance to the onset of the assumed age-related changes that are addressed by the specific preparation intentions, indicating a discounting of delayed outcomes (Green & Myerson, 2004). Accordingly, preparation for third and fourth age were strongest among those age groups that were about to encounter the challenges that are tied to the notion of third and fourth age, respectively. It should be noted, however, that the cross-sectional nature of our study does not allow us to disentangle age-related changes and cohort effects. It would thus be interesting to investigate longitudinal changes in levels of preparation and planning, in order to see whether the steep increase in preparation and planning that was found in our study across age groups is replicated in similar age-related changes across time.

Similarly, it is also possible to explain the observed domain differences in terms of a cohort effect. Different socialization experiences might explain why older cohorts should appraise certain life domains more than others whereas younger cohorts show a different profile of preferences regarding the importance of life domains (e.g., financial security and taking responsibility for emergencies may be a more highly valued goal for the older birth cohorts, whereas younger generations focus more on activities). We are less convinced by this alternative explanation of our findings, however, because previous studies on values and goals did not reveal any strong evidence for such a pattern of cohort differences with regard to values and goals (e.g. Brandtstädter, Renner, & Baltes-Götz, 1989; Brandtstädter et al., 2010). Furthermore, a recent longitudinal study by Mak and Sörensen (2012) showed that different facets of preparation for care needs actually increased over a 6.5-year interval in a sample of adults aged 70 years or older with increased risk of Alzheimer’s disease. Nevertheless, further longitudinal studies on changes in preparation intentions are clearly needed in order to uniquely identify effects of age and cohort on the different facets of preparation and also to further validate the assumed distinction of third- and fourth-age preparations.

Assessment of preparation: General levels of domain-specific preparation vs. specific behaviours

Our questionnaire assesses preparation for age-related changes in a way that enables us to ask participants about general levels of preparation. It covers a wide range of diverse behaviours that are conducted with the intention to prepare for old age and age-related changes with respect to a variety of life domains. Preparation intentions were assessed in a generic format, that is, by assessing the personal involvement with respect to a larger set of different possible activities that serve the same purpose of preparing oneself for age-related changes in a specific life domain. Specific examples of such activities were given, but the amount of involvement in these activities had to be stated in the form of summary ratings. This form of assessing preparation deviates from already existing measures of preparation measuring current or previous engagement in specific concrete behaviours or activities that are assumed to reflect preparation.

Both approaches have advantages and shortcomings. A major advantage of the generic approach is that it explicitly specifies the function or underlying motive of preparation activities in the assessment procedure (i.e., the intention to prepare oneself for old age in a certain life domain). In order to count as preparation for age-related changes, an activity or behaviour has to be conducted with the aim to enhance living conditions in old age, which is what defines an activity as preparation. In the absence of such an intention to improve the conditions of one’s own aging, the very same activity would not count as a case of preparation, but would have to be categorized differently. Focusing on specific behavioural activities only, without explicitly specifying their underlying motivations thus runs the risk of confounding preparation with other forms of behaviour that serve a different function.⁴ Secondly, the very same preparation goal can typically be pursued by different behavioural activities. Which of these behaviours an individual performs, strongly depends on his or her abilities and opportunities, as well as on individual risk factors and life circumstances. Most importantly, different types of preparation activities may vary in effectiveness depending on a person’s age. Thus, in order to compare the amount of preparation in a domain across different age groups, it is important that preparation is assessed at an aggregate level comprising different forms of preparation (mental and behavioural) rather than specific behaviours, in order not to confound age differences in preparation motivation with differences in the age-dependent instrumentality of specific behaviours for these motivations. Thirdly, the generic format allows for a parsimonious and straightforward assessment of levels of domain-specific preparation activities compared to answering long lists of behavioural activities. It cannot be denied that some practical and theoretical purposes require an assessment of specific and concrete behavioural preparation activities rather than focusing on generic preparation intentions. It thus may sometimes be necessary to measure absolute engagement in specific behavioural indicators directly.

When evaluating the practical value of different forms of assessing preparation for old age, it is interesting to think of specific contexts in which an assessment of domain-specific preparation activities might be of interest. For example, such an assessment (e.g. in a doctor’s office, in a counseling centre) can help to identify people who are at risk of ending up being insufficiently prepared for age-related health risks or in the financial domain. Alternatively, large companies might use such an instrument to be informed of the plans and levels of preparation of their older employees with regard to work-related issues (e.g. with regard to vocational training, retirement plans). Relatedly, in the context of urban management, levels of domain-specific preparation may be assessed in a survey or poll to find out what is on people’s minds with regard to age-related changes, and in order to identify specific domains for which an improved infrastructure should be provided (e.g. housing might be the dominant issue, or mobility and public transport, or leisure activities). In most of these cases, assessing general levels of preparation allows for a comprehensive and parsimonious assessment of the relative engagement in different life domains, and seems to be a suitable approach. However, in order to validate our assumption and the usability of our questionnaire in predicting actual behaviours, assessing and relating both constructs is a necessary further step to better understand the relationship and shortcomings as well as the advantages of both approaches, especially since actual preparation behaviour seems not to be widely performed by all people. Investigating when and how intentions are translated into actual behaviours is clearly a challenge for further research.

Implications for further research

Our findings show that preparation for age-related changes is “on people’s minds” throughout the adult life span, but they also show that preparation for old age might be a matter of temporal distance: Younger age groups showed considerably lower levels of preparation than older participants, which might at least partly be explained by a lack of immediate or urgent age-related concerns. In this regard, it would be immensely interesting to disentangle the effects of historical events and their interactions with age and cohort effects on levels of preparation for old age with a cross-sequential design or by comparing levels of old age preparation in different cultures, which would allow us to evaluate the effects that expected demographic and social policy changes have on old age preparations for different cohorts. For example, how do people of different age groups cope with political announcements to reduce efforts of the government to provide financial and social security for older persons, combined with strong demands for self-provision? It would be interesting to investigate whether these “natural interventions” affect levels of preparation differently for young, middle-aged, and older people and also might have differential effects on specific dimensions of old age preparation.

With this regard, one major limitation of our study has to be kept in mind. Even though we recruited a large, age-heterogeneous sample via local registry offices (supplemented by a convenience sample), our sample was not designed to be representative of the entire German population. Therefore, more generalized implications have to be interpreted with caution, and larger studies incorporating representative samples including more persons (e.g. with lower socioeconomic status) would be desirable in order to make assumptions that can be generalized to the population.

Replicating previous findings (Prenda & Lachman, 2001), personality variables were differentially related to reporting more preparation activities. In particular, conscientiousness seems to be a personality trait that fosters the initiation of preparatory activities. Furthermore, extraversion and openness to experience were specifically related to third-age preparations, strengthening the assumption of two distinct concepts, with active third-age preparations being related to making the best of one’s young old years for example by lifelong learning, keeping active and being socially integrated. However, relationships were small to medium-sized, indicating that there might be other factors that explain why some people prepare more than others in various domains.

Another interesting line of research would thus be to identify further determinants of preparation such as age-related goals and expectations, as they are, for example, conceptualized in age stereotypes and views on one’s own aging. Such expectations for the future have been demonstrated to be crucial for the interpretation of age-related changes, for changes in self-views across the life span, as well as for age-related action regulation in a variety of domains (Hoppmann, Gerstorf, Smith, & Klumb, 2007; Kornadt & Rothermund, 2011b, 2012; Levy, 2009; Rothermund & Brandtstädter, 2003a; Wurm, Tomasik, & Tesch-Römer, 2010). However, their relationship to preparation for age-related changes has not yet been investigated. A more thorough investigation of the link between views on aging and levels of preparation for old age seems promising, especially since both variables have been shown to be context-dependent and domain-specific (Casper et al., 2011; Kornadt & Rothermund, 2011a). Domain-specific age stereotypes and views on one’s own aging may provide an important determinant for those aging-related hopes and fears that give rise to corresponding preparation intentions. A related line of research could focus on associations between life-domains and gain vs. loss expectations that should promote specific types of goals and motivations (promotion vs. prevention or maintenance goals) that have already been shown to play a major role in explaining age differences in developmental regulation processes (Ebner, Freund, & Baltes, 2006). Investigating individual differences in domain-specific expectations, goals, and views on aging thus seems a promising avenue for an understanding of individual differences in the quality and quantity of preparation for old age, and might also lead to the development of effective interventions on the individual and the societal level.

Outlook

Western societies have to face demographic changes and progressive aging of their populations in the upcoming years. By implication, people are going to spend more years of their life in older age or retirement (Adams & Rau, 2011). Those expected demographic changes have recently been linked to future scenarios in which societal support for old people breaks down or is strongly reduced, highlighting the demand for individual preparations regarding provision and care in old age. A thorough investigation of individual levels of preparation and planning for old age and aging across the adult life span is thus of considerable societal interest.

In times where biographies, especially in the latter part of the lifespan, are less structured by external regulations and norms (Freund, Nikitin, & Ritter, 2009), and in which longevity is ever-increasing, it becomes more of a personal responsibility to manage and structure one’s life in old age (Ekerdt, 1998; Street & Desai, 2011). Therefore, having thought about and being prepared for at least some of the changes that await a person during the aging process, seems to be beneficial for older persons. Investigating preparation with a multidimensional, life-span approach might thus contribute to the understanding and promotion of “successful” aging.

Footnotes

Acknowledgements

We thank Clemens Tesch-Römer for valuable comments in discussing the outline of this work, and Carina Giesen for valuable comments on an earlier draft of the manuscript. We also thank Karoline Bading for her help in collecting the data.

Funding

This work was supported by two grants of the VolkswagenStiftung (Az. II / 83 142, Az. 86 758) to Klaus Rothermund.

Notes

References

Adams

G. A.

Rau

B. L.

(2011). Putting off tomorrow to do what you want today: Planning for retirement. American Psychologist, 66, 180–192. doi:10.1037/a0022131

Baltes

P. B.

(1997). On the incomplete architecture of human ontogeny: Selection, optimization, and compensation as foundation of developmental theory. American Psychologist, 52, 366–380. doi:10.1037/0003-066X.52.4.366

Baltes

P. B.

Freund

A. M.

(2003). The intermarriage of wisdom and selective optimization with compensation: Two meta-heuristics guiding the conduct of life. In Keyes

C. L. M.

Haidt

(Eds.), Flourishing: Positive psychology and the life well lived (pp. 249–273). Washington, DC: American Psychological Association.

Baltes

P. B.

Glück

Kunzmann

(2002). Wisdom: Its structure and function in regulating successful life span development. In Snyder

C. R.

Lopez

S. J.

(Eds.), Handbook of positive psychology (pp. 327–347). New York, NY: Oxford University Press.

Baltes

P. B.

Smith

(2003). New frontiers in the future of aging: From successful aging of the young old to the dilemmas of the fourth age. Gerontology, 49, 123–135. doi:10.1159/000067946

Brandtstädter

Renner

Baltes-Götz

(1989). Entwicklung von Wertorientierungen im Erwachsenenalter: Quersequentielle Analysen [Development of values in adulthood: Cross-sequential analyses]. Zeitschrift für Entwicklungspsychologie und Pädagogische Psychologie, 21, 3–23.

Brandtstädter

Rothermund

(2002). The life course dynamics of goal pursuit and goal adjustment: A two process framework. Developmental Review, 22, 117–150. doi:10.1006/drev.2001.0539

Brandtstädter

Rothermund

(2003). Intentionality and time in human development and aging: Compensation and goal adjustment in changing developmental contexts. In Staudinger

U. M.

Lindenberger

(Eds.), Understanding human development: Dialogues with life-span psychology (pp. 103–124). Dordrecht, The Netherlands: Kluwer.

Brandtstädter

Rothermund

Kranz

Kühn

(2010). Final decentrations: Personal goals, rationality perspectives, and the awareness of life’s finitude. European Psychologist, 15, 152–163. doi:10.1027/1016-9040/a000019

10.

Casper

Rothermund

Wentura

(2011). The activation of specific facets of age stereotypes depends on individuating information. Social Cognition, 29, 393–414. doi:10.1521/soco.2011.29.4.393

11.

Dempster

A. P.

Laird

N. M.

Rubin

D. B

. (1977). Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society. Series B (Methodological), 39, 1–38. Retrieved from https://www-jstor-org.web.bisu.edu.cn/stable/2984875

12.

Denton

M. A.

Kemp

C. L.

French

Gafni

Joshi

Rosenthal

C. J.

Davies

(2004). Reflexive planning for later life. Canadian Journal on Aging, 23, S71–S82. doi:10.1353/cja.2005.0031

13.

Donaldson

Earl

J. K.

Muratore

A. M.

(2010). Extending the integrated model of retirement adjustment: Incorporating mastery and retirement planning. Journal of Vocational Behavior, 77, 279–289. doi:10.1016/j.jvb.2010.03.003

14.

Ebner

N. C.

Freund

A. M.

Baltes

P. B.

(2006). Developmental changes in personal goal orientation from young to late adulthood: From striving for gains to maintenance and prevention of losses. Psychology and Aging, 21, 664–678. doi:10.1037/0882-7974.21.4.664

15.

Ekerdt

D. J

. (1998). The expected life course: Often unfulfilled, often consequential. Institute on Aging Distinguished Lecture Series. University of North Carolina Institute on Aging. Retrieved from http://www.aging.unc.edu/infocenter/resources/1998/ekerdtd.pdf

16.

Ekerdt

D. J.

(2004). Born to retire: The foreshortened life course. Gerontologist, 44, 3–9. doi:10.1093/geront/44.1.3

17.

Ekerdt

D. J.

Hackney

Kosloski

DeViney

(2001). Eddies in the stream. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 56, S162–S170. doi:10.1093/geronb/56.3.S162

18.

Ekerdt

D. J.

Kosloski

DeViney

(2000). The normative anticipation of retirement by older workers. Research on Aging, 22, 3–22. doi:10.1177/0164027500221001

19.

Freund

A. M.

Baltes

P. B.

(1998). Selection, optimization, and compensation as strategies of life management: Correlations with subjective indicators of successful aging. Psychology and Aging, 13, 531–543. doi:10.1037/0882-7974.13.4.531

20.

Freund

A. M.

Nikitin

Ritter

J. O.

(2009). Psychological consequences of longevity: The increasing importance of self-regulation in old age. Human Development, 52, 1–37. doi:10.1159/000189213

21.

Gluth

Ebner

N. C.

Schmiedek

(2010). Attitudes toward younger and older adults: The German aging semantic differential. International Journal of Behavioral Development, 34, 147–158. doi:10.1177/0165025409350947

22.

Graefe

van Dyk

Lessenich

(2011). Altsein ist später. Alter(n)snormen und Selbstkonzepte in der zweiten Lebenshälfte [Being old is occurring later. Age-related norms and self-concepts in the second half of life]. Zeitschrift für Gerontologie und Geriatrie, 44, 299–305. doi:10.1007/s00391-011-0190-5

23.

Green

Myerson

(2004). A discounting framework for choice with delayed and probabilistic rewards. Psychological Bulletin, 130, 769–792. doi:10.1037/0033-2909.130.5.769

24.

Heckhausen

Dixon

R. A.

Baltes

P. B.

(1989). Gains and losses in development throughout adulthood as perceived by different adult age groups. Developmental Psychology, 25, 109–121. doi:10.1037/0012-1649.25.1.109

25.

Hershey

D. A.

Brown

C. E.

Jacobs-Lawson

J. M.

Jackson

(2001). Retirees’ perceptions of important retirement planning decisions. The Southwest Journal on Aging, 16, 91–100.

26.

Hershey

D. A.

Henkens

Van Dalen

H. P.

(2007). Mapping the minds of retirement planners: A cross-cultural perspective. Journal of Cross-Cultural Psychology, 38, 361–382. doi:10.1177/0022022107300280

27.

Hira

T. K.

Rock

W. L.

Loibl

(2009). Determinants of retirement planning behaviour and differences by age. International Journal of Consumer Studies, 33, 293–301. doi:10.1111/j.1470-6431.2009.00742.x

28.

Hoppmann

C. A.

Gerstorf

Smith

Klumb

P. L.

(2007). Linking possible selves and behavior: Do domain-specific hopes and fears translate into daily activities in very old age? The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 62B, P104–P111.

29.

Jacobs-Lawson

J. M.

Hershey

D. A.

Neukam

K. A

. (2004). Gender differences in factors that influence time spent planning for retirement. Journal of Women & Aging, 16, 55–69. doi:10.1300/J074v16n03_05

30.

Kornadt

A. E.

Rothermund

(2011a). Contexts of aging: Assessing evaluative age stereotypes in different life domains. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 66B, P547–P556. doi:10.1093/geronb/gbr036

31.

Kornadt

A. E.

Rothermund

(2011b). Dimensionen und Deutungsmuster des Alterns – Vorstellungen vom Altern, Altsein und der Lebensgestaltung im Alter [Dimensions and interpretative patterns of aging – attitudes towards aging, being old, and ways of living in old age]. Zeitschrift für Gerontologie und Geriatrie, 44, 291–298. doi:10.1007/s00391-011-0192-3

32.

Kornadt

A. E.

Rothermund

(2012). Internalization of age stereotypes into the self-concept via future self-views: A general model and domain-specific differences. Psychology and Aging, 27, 164–172. doi:10.1037/a0025110

33.

Lang

F. R.

Baltes

P. B.

Wagner

G. G.

(2007). Desired lifetime and end-of-life desires across adulthood from 20 to 90: A dual-source information model. The Journals of Gerontology: Series B: Psychological Sciences and Social Sciences, 62B, P268–P276. doi:10.1093/geronb/62.5.P268

34.

Lang

F. R.

John

Lüdtke

Schupp

Wagner

G. G.

(2011). Short assessment of the Big Five: Robust across survey methods except telephone interviewing. Behavior Research Methods, 43, 548–567. doi:10.3758/s13428-011-0066-z

35.

Laslett

(1989). A fresh map of life: The emergence of the third age. London, UK: Weidenfeld and Nicholson.

36.

Levy

(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

37.

MacEwen

K. E.

Barling

Kelloway

E. K.

Higginbottom

S. F.

(1995). Predicting retirement anxiety: The roles of parental socialization and personal planning. The Journal of Social Psychology, 135, 203–213. doi:10.1080/00224545.1995.9711424

38.

Mak

Sörensen

(2012). Trajectories of preparation for future care among first-degree relatives of Alzheimer’s disease patients: An ancillary study of ADAPT. The Gerontologist, 52, 531–540. doi:10.1093/geront/gnr119

39.

Muratore

A. M.

Earl

J. K.

(2010). Predicting retirement preparation through the design of a new measure. Australian Psychologist, 45, 98–111. doi:10.1080/00050060903524471

40.

Noone

J. H.

Stephens

Alpass

(2010). The Process of Retirement Planning Scale (PRePS): Development and validation. Psychological Assessment, 22, 520–531. doi:10.1037/a0019512

41.

Noone

J. H.

Stephens

Alpass

F. M.

(2009). Preretirement planning and well-being in later life. Research on Aging, 31, 295–317. doi:10.1177/0164027508330718

42.

Petkoska

Earl

J. K.

(2009). Understanding the influence of demographic and psychological variables on retirement planning. Psychology and Aging, 24, 245–251. doi:10.1037/a0014096

43.

Pinquart

Sörensen

(2000). Entwicklung eines Fragebogens zur Vorbereitung von Senioren auf einen zukünftigen Pflegebedarf (FVP). Skalenkonstruktion und teststatistische Überprüfung [Development of a questionnaire about preparation for future care needs (PFCN): Scale construction and psychometric properties]. Diagnostica, 46, 130–139. doi:10.1026//0012-1924.46.3.130

44.

Pinquart

Sörensen

(2002). Psychological outcomes of preparation for future care needs. Journal of Applied Gerontology, 21, 452–470. doi:10.1177/073346402237632

45.

Prenda

K. M.

Lachman

M. E.

(2001). Planning for the future: A life management strategy for increasing control and life satisfaction in adulthood. Psychology and Aging, 16, 206–216. doi:10.1037//0882-7974.16.2.206

46.

Rammstedt

(2007). Who worries and who is happy? Explaining individual differences in worries and satisfaction by personality. Personality and Individual Differences, 43, 1626–1634. doi:10.1016/j.paid.2007.04.031

47.

Rothermund

Brandtstädter

(2003a). Age stereotypes and self-views in later life: Evaluating rival assumptions. International Journal of Behavioral Development, 27, 549–554. doi:10.1080/01650250344000208

48.

Rothermund

Brandtstädter

(2003b). Coping with deficits and losses in later life: From compensatory action to accommodation. Psychology and Aging, 18, 896–905. doi:10.1037/0882-7974.18.4.896

49.

Schindler

Staudinger

U. M.

(2008). Obligatory and optional personal life investments in old and very old age: Validation and functional relations. Motivation & Emotion, 32, 23–36. doi:10.1007/s11031-007-9078-5

50.

Schindler

Staudinger

U. M.

Nesselroade

J. R.

(2006). Development and structural dynamics of personal life investment in old age. Psychology and Aging, 21, 737–753. doi:10.1037/0882-7974.21.4.737

51.

Sörensen

Duberstein

P. R.

Chapman

Lyness

J. M.

Pinquart

(2008). How are personality traits related to preparation for future care needs in older adults? Journals of Gerontology. Series B: Psychological Sciences, 63B, P328–P336.

52.

Sörensen

Mak

Chapman

Duberstein

P. R.

Lyness

J. M.

(2012). The relationship of preparation for future care to depression and anxiety in older primary care patients at 2-year follow-up. The American Journal of Geriatric Psychiatry, 20, 887–894. doi:10.1097/JGP.0b013e31822ccd8c

53.

Sörensen

Pinquart

(2001). Development a measure of older adults’ preparation for future care needs. The International Journal of Aging & Human Development, 53(2), 137–165. doi:10.2190/1r0d-30tc-f4k1-f0dw

54.

Stawski

R. S.

Hershey

D. A.

Jacobs-Lawson

J. M.

(2007). Goal clarity and financial planning activities as determinants of retirement savings contributions. The International Journal of Aging & Human Development 64, 13–32. doi:10.2190/13GK-5H72-H324-16P2

55.

Street

Desai

(2011). Planning for old age. In Settersten

R. A.

Angel

J. L.

(Eds.), Handbook of Sociology of Aging (pp. 379–397). New York, NY: Springer.

56.

Topa

Moriano

J. A.

Depolo

Alcover

C.-M.

Morales

J. F.

(2009). Antecedents and consequences of retirement planning and decision-making: A meta-analysis and model. Journal of Vocational Behavior, 75, 38–55. doi:10.1016/j.jvb.2009.03.002

57.

Tornstam

(1997). Gerotranscendence: The contemplative dimension of aging. Journal of Aging Studies, 11, 143–154. doi:10.1016/S0890-4065(97)90018-9

58.

van Dyk

Lessenich

Denninger

Richter

(2013). Gibt es ein Leben nach der Arbeit? Zur diskursiven Konstruktion und sozialen Akzeptanz des “aktiven Alters” [Is there a life after work? On the discoursive construction and social acceptance of the “active age”]. WSI-Mitteilungen, 5, 321–228.

59.

Wang

Shultz

K. S.

(2009). Employee retirement: A review and recommendations for future investigation. Journal of Management, 36, 172–206. doi:10.1177/0149206309347957

60.

Wurm

Tesch-Römer

Tomasik

M. J.

(2010). On the importance of a positive view on ageing for physical exercise among middle-aged and older adults: Cross-sectional and longitudinal findings. Psychology and Health, 25, 25–42. doi:10.1080/08870440802311314