The Relation Between Age and Three-Dimensional Wisdom: Variations by Wisdom Dimensions and Education

The Relation Between Age and Three-Dimensional Wisdom: Variations by Wisdom Dimensions and Education Abstract Objectives Curvilinear relations between age and wisdom have been found in prior research with its peak either in young adulthood or in midlife. This study tested whether the association between age and three-dimensional wisdom differed for the cognitive, reflective, and compassionate (affective) wisdom dimensions and whether the results varied by education. Method OLS regression was used to analyze an online sample of 14,248 adults between the ages of 18 and 98 years (mean [M] = 36.46, standard deviation [SD] = 12.68) from four educational groups. Results The relation between age and three-dimensional wisdom was curvilinear and varied in shape by wisdom dimensions and education. The association between age and wisdom followed an inverse U-curve with the peak at midlife and almost opposing shapes for the cognitive and compassionate dimensions and an intermediate shape for the reflective dimension. Education was positively related to wisdom and affected the shape of the relationships. Discussion Because the association between age and wisdom appears to be curvilinear and varies by education, studies that test a linear association without separating the sample by education might arrive at the wrong conclusions. In general, education seems to promote the attainment of wisdom and also inoculate individuals against a decline in wisdom later in life. Successful aging, Human development, Gender It is often assumed colloquially that wisdom comes with age and experience, yet empirically and anecdotally this is not necessarily the case (Glück, 2016). Although life expectancy has increased during the last century (Murphy, Xu, & Kochanek, 2013), it appears that wisdom has not. In fact, wisdom seems to be rare among any age group, including the old. Life experiences might be a necessary but not sufficient condition for the development of wisdom. Instead, most experts and lay people agree that learning from life experiences is a prerequisite for the emergence of wisdom (Staudinger & Glück, 2011). Hence, if wisdom depends on the willingness to learn from experiences it might indeed increase with age but only up to a point until people believe that they have learned enough about life. If people become dogmatic in their views, think that they know more than they do, and do not realize the limits and uncertainty of knowledge (Kitchener & Brenner, 1990), wisdom might actually decrease with age. This might explain why a curvilinear association in the form of an inverse U-curve has been found between age and wisdom with the apex in middle age (Bergsma & Ardelt, 2012; Thomas, Bangen, Ardelt, & Jeste, 2017; Webster, Westerhof, & Bohlmeijer, 2014). Moreover, wisdom is a multifaceted concept that consists of cognitive, reflective, and affective elements (Ardelt, 2000; Blanchard-Fields & Norris, 1995; Kramer, 1990; Takahashi & Overton, 2002). Although the cognitive or knowledge aspect of wisdom might peak and then decline after a certain age because of an increase in rigidity or dogmatism (Schultz & Searleman, 2002), the reflective and emotional elements might not decrease but might even increase with age due to psychosocial growth (Erikson, 1963) and increases in perspective-taking skills, emotional regulation, and empathic concern or compassion for others (Chopik, O’Brien, & Konrath, 2017; Neff & Pommier, 2013; Scheibe & Carstensen, 2010). Individuals with a higher education are longer exposed to learning and might therefore also be more open and motivated to learn from life and develop greater wisdom. Hence, in addition to a positive association between education and various wisdom measures (Ardelt, 2003; Bergsma & Ardelt, 2012; Cheraghi, Kadivar, Ardelt, Asgari, & Farzad, 2015; Glück et al., 2013; Webster et al., 2014), the relation between age and wisdom and its individual elements might also depend on education. It is possible that the peak in the curvilinear association between age and wisdom comes earlier for those with a lower education or that wisdom does not decline with age for those with a higher education. The purpose of the current study was to examine whether the association between age and wisdom varied for the cognitive, reflective, and compassionate (affective) dimensions of the three-dimensional wisdom scale (Ardelt, 2003) and for different education levels, using a large German internet survey. The Concept of Wisdom Many definitions of wisdom exist (Staudinger & Glück, 2011). For example, wisdom has been defined as expert knowledge in the fundamental pragmatics of life (Baltes & Staudinger, 2000), orienting knowledge about what is good and right to live well (Fischer, 2015), “deep accurate insight and understanding of oneself and the central existential issues of life, plus skillful benevolent responsiveness” (Walsh, 2015, p. 282), and self-transcendence (Levenson, Jennings, Aldwin, & Shiraishi, 2005). We define and operationalize wisdom as the integration of cognitive, reflective, and compassionate personality characteristics (Ardelt, 2003) based on Clayton and Birren’s (1980) earlier research. The cognitive dimension describes a wise person’s desire to know a deeper truth and the ability to thoroughly understand the intrapersonal and interpersonal aspects of life, which includes an understanding and acceptance of the ultimate limits of knowledge, the positive and negative aspects of human nature, and life’s unpredictability and uncertainties. The reflective dimension refers to a wise person’s ability to perceive phenomena and events from multiple perspectives and to avoid subjectivity and projections, which requires self-examination, self-awareness, and self-insight and a reduction in self-centeredness. Diminished self-centeredness, in turn, is likely to result in greater understanding of life in general and other people in particular, leading to greater tolerance and sympathetic and compassionate love for others, which represents the compassionate dimension of wisdom. Although wisdom is defined as the integration of the three dimensions, greater insight into the overall association between age and wisdom might be gained by analyzing the relation between age and the three dimensions separately. Age and Wisdom Sternberg (2005) discussed five theoretical views of the age and wisdom relationship. According to the “received” view, which is often shared by laypeople, wisdom comes with age. That is, wisdom is assumed to be low or nonexistent earlier in life until it is received in old age. The “fluid intelligence” view suggests that wisdom is similar to fluid intelligence, or the “ability to think flexibly in novel ways” (Sternberg, 2005, p. 13) and, therefore, tends to increase until early adulthood, then levels off, until it starts to decline in middle age (Kievit, Davis, Griffiths, Correia, & Henson, 2016). The “crystallized intelligence” view describes wisdom as similar to crystallized intelligence or the ability to use acquired knowledge, experience, and skills effectively. Based on this view, wisdom increases continuously up to old age until it starts to decline, on average, about 10 years before one’s death due to the onset of disease, illness, and terminal decline (Kaufman, 2001). A fourth view combines the “fluid and crystallized intelligence” view. Wisdom is theorized to increase early in life until middle age when crystallized intelligence continues to rise while fluid intelligence begins to decline. This decrease in fluid intelligence results in a net negative effect on wisdom, so that wisdom increases until late middle age and then declines. Classical empirical work by Horn and Cattell (1966) showed that aspects of fluid intelligence such as associative memory, intellectual speed, and intellectual level, tended to decline considerably after early adulthood. By contrast, aspects of crystallized intelligence, including mechanical knowledge, ideational and associational fluency, and verbal comprehension, tended to increase with age but leveled off by midlife. These results have been replicated in later research with performance (fluid) intelligence and verbal (crystallized) intelligence (e.g., Kaufman, 2001). Hence, marked decreases in fluid intelligence and smaller gains in crystallized abilities by midlife predict an overall downward trend starting in late midlife for the “fluid and crystallized intelligence” view. Finally, the “declining wisdom” view advocated by Meacham (1990) states that wisdom declines with age, starting in early life. According to this view, children are considered wise and open to the world, but hardships and successes throughout life tend to reduce this wisdom. Hardships might diminish wisdom if they result in doubt and a loss of meaning and confidence in the world, whereas personal successes might decrease wisdom if they lead to a self-centered certainty and over-confidence in the world accompanied by dogmatism, prejudice, and intolerance. Three of the five views assume that wisdom increases during the early years of life due to normative personality and intellectual development and the exposure to education (Richardson & Pasupathi, 2005), which has been supported empirically (Ferrari, Kahn, Benayon, & Nero, 2011; Pasupathi, Staudinger, & Baltes, 2001). However, after an initial maturationah age in adulthood. Yet, older adults have a higher probability to be wiser than younger individuals because growth in wisdom is assumed to take time (Kekes, 1983). This means that wisdom might increase with age only for those individuals who actively pursue its acquisition and are able to learn from life experiences (Staudinger & Glück, 2011 increase in wisdom from childhood to adolescence and young adulthood, wisdom might not continue to develop witef>). Empirical evidence has shown that in adulthood, age is not necessarily linearly correlated with wisdom (Baltes & Staudinger, 2000; Glück et al., 2013; Levenson et al., 2005; Webster, 2007), although several studies have found a curvilinear association between age and wisdom, assessed by the Three-Dimensional Wisdom scale (3D-WS; Ardelt, 2003), integrating cognitive, reflective, and compassionate dimensions, and the Self-Assessed Wisdom Scale (SAWS; Webster, 2007), which is the average of critical life experiences, reminiscence/reflectiveness, openness to experience, emotional regulation, and humor, with an increase in wisdom until midlife followed by a decrease (Bergsma & Ardelt, 2012; Thomas et al., 2017; Webster et al., 2014). However, not all dimensions of wisdom might decline with age. While cognitive aspects of wisdom might follow the “fluid intelligence” view with an increase in wisdom until midlife and a decrease thereafter, noncognitive elements of wisdom might actually increase with age. For example, according to socioemotional selectivity theory (Carstensen & Mikels, 2005; Charles & Carstensen, 2009; Scheibe & Carstensen, 2010), older adults might be less interested in gaining greater knowledge but more likely than younger adults to engage in emotional regulation to maximize positive affect and minimize negative affect in an effort to make the best of their remaining time. One way to diminish negative affect is to perceive the larger picture, focus on the positive aspects, and comprehend others’ perspectives, which tends to decrease self-centeredness and increase compassion for others. Indeed, research has shown that age is positively related to emotional regulation, perspective-taking skills, and empathic concern or compassion for others (Chopik et al., 2017; Neff & Pommier, 2013; Scheibe & Carstensen, 2010). Research also has found that a motivation for psychosocial growth and favorable psychological, social, and historical conditions might foster the development of wisdom (Baltes & Smith, 2008). For example, among German adults between the ages of 20 and 87, age was positively related to wisdom-related knowledge for those with higher levels of moral reasoning but not for those with moral reasoning scores below the median (Pasupathi & Staudinger, 2001). In an Iranian sample of adults between 18 and 80 years of age, age was positively correlated with three-dimensional wisdom and the reflective and compassionate wisdom dimensions for those with a higher education but unrelated for those with a lower education. Yet, age and the cognitive wisdom dimension were unrelated among the higher educated participants and were negatively correlated among those with a lower education (Cheraghi et al., 2015). Moreover, in a study comparing college students (mean [M] = 21 years) with older adults between ages 52 and 87 (M = 71 years), college-educated older adults had higher average three-dimensional wisdom scores and scores on the reflective and compassionate wisdom dimensions than young college students, but older adults without a college degree had significantly lower average wisdom scores and scores on the cognitive wisdom dimension than college-educated older adults and college students (Ardelt, 2010). In longitudinal research, Wink and Helson (1997) found that highly educated participants, and clinical psychologists in particular, experienced gains in practical wisdom from age 27 to 52. Taken together, these studies support the “crystallized intelligence” view for the noncognitive dimensions of wisdom for people who are highly educated and/or have the motivation to grow in wisdom. Education might provide the “luxury” to continue to be curious about the world, one’s place in it, and the deeper meaning of life. Adults with a lower education and employment that is characterized by physical labor and/or repetitive work might be too exhausted to engage in activities of psychosocial growth. Research by Kohn and colleagues (Kohn & Schooler, 1983; Kohn et al., 2000) demonstrated that individuals with a higher education were more likely to endorse autonomy, self-directedness, and openness to change, which was reflected and reinforced in their work environment, whereas those with a lower education tended to hold authoritarian, conformative beliefs and were employed in less complex and more supervised, standardized jobs. This resembles Staudinger and Kunzmann’s (2005) distinction between personality growth and personality adjustment. Wisdom is an indicator of personality growth, which requires openness to experiences and a questioning attitude toward societal norms rather than conforming and adjusting to the prevailing societal standards. Therefore, a higher education facilitates personality growth, while a lower education encourages personality adjustment, particularly through work requirements. Hence, a positive association between age and wisdom might be more likely among highly educated individuals than among individuals with a lower education. Yet, those differences and especially nonlinear differences by educational attainment might not be visible in small samples with participants from diverse educational backgrounds. The Present Study The goal of the present study was twofold. First, we examined how age was related to three-dimensional wisdom in adulthood and whether this relation differed for the three individual dimensions. Second, we analyzed whether the associations between age and wisdom and the three individual dimensions varied by educational groups. While the five theoretical views of the age and wisdom relationship assume a unidimensional conception of wisdom, we predicted that the associations between age and the three individual wisdom dimensions would not follow the same pattern. We hypothesized that the relation between age and composite three-dimensional wisdom in adulthood followed an inverse U-curve with the peak at midlife (Hypothesis 1), based on prior studies (Bergsma & Ardelt, 2012; Thomas et al., 2017; Webster et al., 2014) and similar to the “fluid and crystallized intelligence” view (Sternberg, 2005). Webster and colleagues (2014, p. 215) found that older adults tend to have more life experiences but also less openness to new ideas, whereas younger adults tend to be less experienced but more accepting of new knowledge, concluding that “midlife adults have a critical mass of life experiences and the cognitive strengths and emotional resources to process such events.” Yet, people at the top of their careers and/or having raised their children to become independent adults might feel that they know how the world works rather than understanding, as Socrates did, that there is much more to know. If individuals do not continue to learn from experiences, they might become dogmatic and unwilling to search for a deeper truth or to engage in reflective thinking and self-reflection to question their existing beliefs (Meacham, 1990; Rokeach, 1954). This curvilinear association might be most pronounced for the cognitive wisdom dimension, which represents a desire to know a deeper truth, with a steeper decline after midlife (Hypothesis 2). The reflective wisdom dimension, which consists of perspective-taking, self-reflection, and the avoidance of blame, might first follow the same shape but might actually increase in old age when individuals use emotional regulation to maximize positive affect (Carstensen & Mikels, 2005; Charles & Carstensen, 2009; Scheibe & Carstensen, 2010) and struggle with the psychosocial crisis of integrity versus despair (Hypothesis 3). According to Erikson’s (1963) stage model of human development, older people can achieve ego integrity and greater wisdom when they engage in life review and reflect on the past as it relates to the present to learn acceptance of the totality of their lives without despairing over past failures or mistakes, missed opportunities, or the nearing of death (Staudinger, 2001). Later life might also provide more time for this kind of self-reflection, particularly after retirement (Ardelt, 2000). The compassionate wisdom dimension requires a reduction in self-centeredness to develop sympathy and compassion for others. This reduction in self-centeredness might begin in midlife and continue into old age if individuals master Erikson’s psychosocial crisis of generativity versus self-absorption or stagnation. During midlife, many adults display an “interest in establishing and guiding the next generation” (Erikson, 1959, p. 97) by focusing more on the well-being of others and contributing to the success and well-being of future generations, not just their own children or grandchildren. Hence, the compassionate wisdom dimension was expected to increase, starting in midlife (Hypothesis 4). Individuals with a higher education are probably more likely to continue to learn new things and, therefore, might be more open and motivated to learn from life experiences than those with a lower education. A higher education is also more likely to result in leadership positions, which makes it easier to engage in generativity. Based on this reasoning, we predicted that the zenith of the curvilinear relation of age on wisdom and the cognitive wisdom dimension would be higher and occur later for individuals with a higher education than for those with a lower education or that age might even be positively related to wisdom and the cognitive dimension for those with a higher education (Hypothesis 5). Similarly, we hypothesized that individuals with a higher education might be more interested in life review and reflection in old age and motivated to engage in generativity in midlife than those with a lower education with increases in the reflective wisdom dimension during later life and in the compassionate wisdom dimension after midlife only expected for the high school and university education groups (Hypothesis 6). In exploratory analyses, we also tested for possible moderation effects of gender on the association between age and/or education on wisdom and its three dimensions. Method Procedure In 2006, the German magazine GEO (equivalent to National Geographic in the United States) published a three-part series about wisdom in their March, April, and May issues and offered their readers the opportunity to take an online “wisdom test.” Participants received an average score for the three-dimensional wisdom scale and its cognitive, reflective, and compassionate dimensions. General interpretations of scores below the midpoint of the scales (a score of <3), slightly above the midpoint (a score between 3 and <4), and in the top of the scale (a score between 4 and 5) were provided by the author of the GEO wisdom article after feedback from the senior author. The interpretations included general statements and tips about self-improvement and emphasized the positive even for those who scored below the midpoint. In addition to wisdom, age, gender, and education were assessed. Readers were informed that the anonymous data would be analyzed in a study. GEO technicians collected the raw data that were submitted between March 17 and May 5, 2006, including time and date of submission. Measures Three-dimensional wisdom was assessed by the cognitive, reflective, and compassionate (affective) dimensions of the 3D-WS (Ardelt, 2003) on two 5-point scales (1 = strongly agree and 5 = strongly disagree or 1 = definitely true of myself and 5 = not true of myself). The 14 items of the cognitive wisdom dimension measured the ability or willingness to understand a situation or phenomenon thoroughly, knowledge of the positive and negative aspects of human nature, an acknowledgement of ambiguity and uncertainty in life, and the ability to make important decision despite life’s unpredictability and uncertainties (e.g., Ignorance is bliss—reversed). The 12 items of the reflective wisdom dimension gauged the ability and willingness to look at phenomena and events from different perspectives and the absence of bitterness, subjectivity, and projections (e.g., I always try to look at all sides of a problem). The 13 items of the compassionate wisdom dimension assessed the presence of positive, caring, and nurturing emotions and behavior and the absence of indifferent or negative emotions and behavior toward others (e.g., Sometimes I feel a real compassion for everyone). All items were scored in the direction of greater wisdom before calculating the average for each wisdom dimension. Cronbach’s α was .72, .77, and .68 for the cognitive, reflective, and compassionate dimensions, respectively, and .74 for the composite wisdom scale consisting of the average of the three dimensions rather than the 39 items (Cronbach’s α was .85 for the 39 items). The cognitive dimension correlated r = .51 with the reflective dimension and r = .45 with the compassionate dimension, and the correlation between the reflective and compassionate wisdom dimensions was r = .48 (all p < .001). Age was measured in years and gender as 0 = male and 1 = female. Education had four categories (1 = elementary education, 2 = intermediate education, 3 = high school education, 4 = university education). To test for nonlinear associations between age and wisdom and all combinations of interaction effects between age, education, and gender on wisdom, age and education were centered at the mean to reduce multicollinearity (Aiken & West, 1991). Sample The original data set contained 28,538 cases. Unfortunately, the GEO technicians did not block multiple submissions from the same IP address nor did they collect the IP addresses of participants, which required a manual clean-up of the data. Initially, we deleted cases with the same demographics, the same score on all wisdom measures, and the same timestamp (to the minute), as these participants likely hit the submit button more than once. All cases that matched exactly on each wisdom item, differed by only one demographic variable, and were entered within 10 minutes of each other were deleted, because we believe curious respondents were trying to determine how changes in the demographic variables influenced the final wisdom score. These procedures led to the removal of 13,309 cases. Subsequently, all cases with missing values on gender, education, and wisdom were removed (234 cases). We also performed an internal validity check on the 3D-WS and deleted 46 cases that strongly endorsed items or strongly disagreed with items that contradicted each other (e.g., strongly endorsing the item “If I see people in need, I try to help them one way or another” and also the items “It’s not really my problem if others are in trouble and need help” and “I often have not comforted another when he or she needed it”). Twenty-one outliers with extremely low or “perfect” 3D-WS scores were also deleted, because they did not appear to reflect valid entries. Finally, we removed 701 participants under the age of 18 or with an age of 99. We were concerned that the nine people who entered 99 did not want to reveal their real age. Minors were deleted for ethical reasons as we did not ask their parents’ permission to participate in the research and because some of the minor ages were clearly fake (e.g., ages of 0–5). After cleaning the data, 14,248 respondents between the ages of 18 and 98 (M = 36.46, standard deviation [SD] = 12.68, median = 35) remained. Approximately half (48.5%) of the participants were women and 7.4% reported having an elementary education (Haupschule), 19.2% an intermediate education (Realschule), 29.5% a high school education (Gymnasium), and 44.0% a university education. The percentage who earn a high school degree (Abitur) in Germany is much lower than in the United States and was only 33% in 2006, compared to 28% who received an elementary education and 39% who obtained an intermediate education (http://www.deutschlandinzahlen.de/tab/deutschland/bildung/schule/allgemeinbildende/schulabsolventen-insgesamt). Given these statistics and considering that not everyone with a high school degree will get a university education, it is clear that the present sample is biased in the direction of a higher education. Results Bivariate Correlations In bivariate correlations, age was unrelated to the 3D-WS, but negatively correlated with the cognitive (r = −.05) and compassionate (r = −.04) wisdom dimensions and positively with the reflective wisdom dimension (r = .07). Education was positively associated with the 3D-WS (r = .16) and with the cognitive (r = .20), reflective (r = .16), and compassionate (r = .04) wisdom dimensions but negatively with age (r = −.09). Being female was positively correlated with the 3D-WS (r = .04), the cognitive (r = .05), and compassionate (r = .10) wisdom dimensions but negatively with the reflective wisdom dimension (r = −.04) and age (r = −.04). Women were less likely than men to report having an elementary education (6% vs. 9%) and a university education (42% vs. 45%) but more likely than men to indicate an intermediate education (21% vs. 18%). All of these correlations were significant at p < .001, although some of the effect sizes were quite small. Multivariate Regression Analyses for the Full Sample To perform the multivariate OLS regression analyses with the full sample, the significance level was set at p ≤.001 due to the large sample size. First, age, gender, and education were entered into the model. Age-square was added if it was significant and then age-cube if it was significant. We also tested whether any of the interaction effects with education and/or gender were statistically significant, using the FORWARD command in SPSS 24. The results for the full sample are displayed in Table 1 and in Figure 1 with the original age range. As predicted in Hypothesis 1, the association between age and wisdom followed an inverse U-curve with the peak at age 43. However, the curve was rather flat as Figure 1 shows. The curvilinear relation was more pronounced for the cognitive wisdom dimension, confirming Hypothesis 2. In fact, a cubic association existed between age and the cognitive wisdom dimension, but with a local minimum at age 17, beyond the current age range, and a maximum at age 42. Figure 1 shows that age tended to be unrelated to the cognitive wisdom dimension until it started to decline after age 42. There was also a cubic relation between age and the reflective wisdom dimension. As expected in Hypothesis 3, the reflective wisdom dimension tended to increase until midlife (age 46), then was relatively stable as depicted in Figure 1, and subsequently tended to increase again after the age of 71. Hypothesis 4 was also supported. The association between age and the compassionate wisdom dimension followed a quadratic U-curve with slight decreases, on average, in the compassionate wisdom dimension until midlife and increases after age 46. However, there was a significant interaction effect between age-squared and education, suggesting that the shape differed by educational group. Because education was centered at the mean (M = 3.10, SD = 0.96), the curvilinear association shown in Figure 1 represents respondents with a high school education the most. The curvilinear shape also applied to those with a university education but with the nadir at age 44. However, for respondents with an elementary education, the shape became an inverse U-curve with the peak at age 37, and for participants with an intermediate education, age appeared to be unrelated to the compassionate wisdom dimension. After controlling for age, education was positively related to the 3D-WS and the cognitive and reflective wisdom dimensions but not the compassionate wisdom dimension. Table 1. Relation Between Age and Three-Dimensional Wisdom Controlling for Education and Gender; OLS Regression Analyses   3D-wisdom  Cognitive dimension  Reflective dimension  Compassionate dimension  Independent variables  b  β  b  β  b  β  b  β  Constant  3.61439    3.75669    3.77487    3.31001    Age  .00093  .03**  .00109  .03  .00337  .09**  −.00179  −.05**  Age2  −.00007  −.04**  −.00001  −.03  −.00022  −.11**  .00009  .05**  Age3  —  —  −.000003  −.06**  .000003  .06**  —  —  Education  .06401  .16**  .09624  .20**  .07926  .16**  .00417  .01  Female (1 = yes)  .03248  .04**  .04024  .04**  −.03547  −.04**  .09173  .10**  Age * Education  —  —  —  —  —  —  −.00092  −.03  Age2 * Education  —  —  —  —  —  —  .00009  .06**  Adjusted R2  .03**  .05**  .04**  .02**    3D-wisdom  Cognitive dimension  Reflective dimension  Compassionate dimension  Independent variables  b  β  b  β  b  β  b  β  Constant  3.61439    3.75669    3.77487    3.31001    Age  .00093  .03**  .00109  .03  .00337  .09**  −.00179  −.05**  Age2  −.00007  −.04**  −.00001  −.03  −.00022  −.11**  .00009  .05**  Age3  —  —  −.000003  −.06**  .000003  .06**  —  —  Education  .06401  .16**  .09624  .20**  .07926  .16**  .00417  .01  Female (1 = yes)  .03248  .04**  .04024  .04**  −.03547  −.04**  .09173  .10**  Age * Education  —  —  —  —  —  —  −.00092  −.03  Age2 * Education  —  —  —  —  —  —  .00009  .06**  Adjusted R2  .03**  .05**  .04**  .02**  Note. N = 14,248; b = unstandardized coefficient; β = standardized coefficient; Age and Education were centered at the mean. ** p ≤ .001. View Large Table 1. Relation Between Age and Three-Dimensional Wisdom Controlling for Education and Gender; OLS Regression Analyses   3D-wisdom  Cognitive dimension  Reflective dimension  Compassionate dimension  Independent variables  b  β  b  β  b  β  b  β  Constant  3.61439    3.75669    3.77487    3.31001    Age  .00093  .03**  .00109  .03  .00337  .09**  −.00179  −.05**  Age2  −.00007  −.04**  −.00001  −.03  −.00022  −.11**  .00009  .05**  Age3  —  —  −.000003  −.06**  .000003  .06**  —  —  Education  .06401  .16**  .09624  .20**  .07926  .16**  .00417  .01  Female (1 = yes)  .03248  .04**  .04024  .04**  −.03547  −.04**  .09173  .10**  Age * Education  —  —  —  —  —  —  −.00092  −.03  Age2 * Education  —  —  —  —  —  —  .00009  .06**  Adjusted R2  .03**  .05**  .04**  .02**    3D-wisdom  Cognitive dimension  Reflective dimension  Compassionate dimension  Independent variables  b  β  b  β  b  β  b  β  Constant  3.61439    3.75669    3.77487    3.31001    Age  .00093  .03**  .00109  .03  .00337  .09**  −.00179  −.05**  Age2  −.00007  −.04**  −.00001  −.03  −.00022  −.11**  .00009  .05**  Age3  —  —  −.000003  −.06**  .000003  .06**  —  —  Education  .06401  .16**  .09624  .20**  .07926  .16**  .00417  .01  Female (1 = yes)  .03248  .04**  .04024  .04**  −.03547  −.04**  .09173  .10**  Age * Education  —  —  —  —  —  —  −.00092  −.03  Age2 * Education  —  —  —  —  —  —  .00009  .06**  Adjusted R2  .03**  .05**  .04**  .02**  Note. N = 14,248; b = unstandardized coefficient; β = standardized coefficient; Age and Education were centered at the mean. ** p ≤ .001. View Large Figure 1. View largeDownload slide Relation between age and three-dimensional wisdom by wisdom dimensions. “____” represents composite three-dimensional wisdom. “............” represents cognitive wisdom dimension. “- - - - - ” represents reflective wisdom dimension. “__ __ __ ” represents compassionate wisdom dimension. Figure 1. View largeDownload slide Relation between age and three-dimensional wisdom by wisdom dimensions. “____” represents composite three-dimensional wisdom. “............” represents cognitive wisdom dimension. “- - - - - ” represents reflective wisdom dimension. “__ __ __ ” represents compassionate wisdom dimension. Multivariate Regression Analyses for Separate Educational Groups The same procedure as for the full sample was followed for the separate analyses by educational group (see Table 2). Because the sample sizes were smaller, the significance level was increased to p <.01. Table 2. Relation between Age and Three-Dimensional Wisdom by Education Controlling for Gender; OLS Regression Analyses Education  Elementary  Intermediate  High school  University  Wisdom Measure  b  β  b  β  b  β  b  β  3D-Wisdom   Constant  3.54827    3.53064    3.59651    3.66183     Age  .00144  .05  .00178  .06  .00025  .01  −.00005  −.00   Age2  −.00024  −.18**  −.00010  −.06*  —  —  —  —   Female (1 = yes)  .00415  .01  .03196  .04  .02574  .03  .03864  .05**   Age * female  —  —  .00318  .07*  —  —  —  —  Cognitive dimension   Constant  3.63719    3.63811    3.75177    3.83827     Age  .00113  .03  .00072  .02  −.00143  −.04  .00072  .02   Age2  −.00033  −.21**  −.00021  −.10**  —  —  −.00015  −.07**   Female (1 = yes)  .00475  .00  .04130  .04  .01388  .02  .06021  .07**   Age * female  —  —  .00477  .09**  —  —  —  —  Reflective dimension   Constant  3.61939    3.70161    3.72336    3.85608     Age  .00333  .09  .00537  .14**  .00411  .10**  .00211  .06*   Age2  −.00018  −.11*  −.00036  −.17**  —  —  −.00027  −.13**   Age3  —  —  .000007  .12*  —  —  .000005  .09*   Female (1 = yes)  .00332  .00  −.05057  −.05*  −.03534  −.04  −.03628  −.04*  Compassionate dimension   Constant  3.38822    3.28081    3.27204    3.32472     Age  −.00014  −.00  .00067  .02  −.00049  −.01  −.00210  −.06**   Age2  −.00021  –.14**  —  —  .00027  .12**  .00012  .06**   Age3  —  —  —  —  −.000005  −.10*  —  —   Female (1 = yes)  .00437  .00  .11296  .13**  .10269  .11**  .09092  .10**  N  1,050  2,736  4,199  6,263  Education  Elementary  Intermediate  High school  University  Wisdom Measure  b  β  b  β  b  β  b  β  3D-Wisdom   Constant  3.54827    3.53064    3.59651    3.66183     Age  .00144  .05  .00178  .06  .00025  .01  −.00005  −.00   Age2  −.00024  −.18**  −.00010  −.06*  —  —  —  —   Female (1 = yes)  .00415  .01  .03196  .04  .02574  .03  .03864  .05**   Age * female  —  —  .00318  .07*  —  —  —  —  Cognitive dimension   Constant  3.63719    3.63811    3.75177    3.83827     Age  .00113  .03  .00072  .02  −.00143  −.04  .00072  .02   Age2  −.00033  −.21**  −.00021  −.10**  —  —  −.00015  −.07**   Female (1 = yes)  .00475  .00  .04130  .04  .01388  .02  .06021  .07**   Age * female  —  —  .00477  .09**  —  —  —  —  Reflective dimension   Constant  3.61939    3.70161    3.72336    3.85608     Age  .00333  .09  .00537  .14**  .00411  .10**  .00211  .06*   Age2  −.00018  −.11*  −.00036  −.17**  —  —  −.00027  −.13**   Age3  —  —  .000007  .12*  —  —  .000005  .09*   Female (1 = yes)  .00332  .00  −.05057  −.05*  −.03534  −.04  −.03628  −.04*  Compassionate dimension   Constant  3.38822    3.28081    3.27204    3.32472     Age  −.00014  −.00  .00067  .02  −.00049  −.01  −.00210  −.06**   Age2  −.00021  –.14**  —  —  .00027  .12**  .00012  .06**   Age3  —  —  —  —  −.000005  −.10*  —  —   Female (1 = yes)  .00437  .00  .11296  .13**  .10269  .11**  .09092  .10**  N  1,050  2,736  4,199  6,263  Note. b = unstandardized coefficient; β = standardized coefficient; Age and Education were centered at the mean. * p < .01. ** p ≤ .001. View Large Table 2. Relation between Age and Three-Dimensional Wisdom by Education Controlling for Gender; OLS Regression Analyses Education  Elementary  Intermediate  High school  University  Wisdom Measure  b  β  b  β  b  β  b  β  3D-Wisdom   Constant  3.54827    3.53064    3.59651    3.66183     Age  .00144  .05  .00178  .06  .00025  .01  −.00005  −.00   Age2  −.00024  −.18**  −.00010  −.06*  —  —  —  —   Female (1 = yes)  .00415  .01  .03196  .04  .02574  .03  .03864  .05**   Age * female  —  —  .00318  .07*  —  —  —  —  Cognitive dimension   Constant  3.63719    3.63811    3.75177    3.83827     Age  .00113  .03  .00072  .02  −.00143  −.04  .00072  .02   Age2  −.00033  −.21**  −.00021  −.10**  —  —  −.00015  −.07**   Female (1 = yes)  .00475  .00  .04130  .04  .01388  .02  .06021  .07**   Age * female  —  —  .00477  .09**  —  —  —  —  Reflective dimension   Constant  3.61939    3.70161    3.72336    3.85608     Age  .00333  .09  .00537  .14**  .00411  .10**  .00211  .06*   Age2  −.00018  −.11*  −.00036  −.17**  —  —  −.00027  −.13**   Age3  —  —  .000007  .12*  —  —  .000005  .09*   Female (1 = yes)  .00332  .00  −.05057  −.05*  −.03534  −.04  −.03628  −.04*  Compassionate dimension   Constant  3.38822    3.28081    3.27204    3.32472     Age  −.00014  −.00  .00067  .02  −.00049  −.01  −.00210  −.06**   Age2  −.00021  –.14**  —  —  .00027  .12**  .00012  .06**   Age3  —  —  —  —  −.000005  −.10*  —  —   Female (1 = yes)  .00437  .00  .11296  .13**  .10269  .11**  .09092  .10**  N  1,050  2,736  4,199  6,263  Education  Elementary  Intermediate  High school  University  Wisdom Measure  b  β  b  β  b  β  b  β  3D-Wisdom   Constant  3.54827    3.53064    3.59651    3.66183     Age  .00144  .05  .00178  .06  .00025  .01  −.00005  −.00   Age2  −.00024  −.18**  −.00010  −.06*  —  —  —  —   Female (1 = yes)  .00415  .01  .03196  .04  .02574  .03  .03864  .05**   Age * female  —  —  .00318  .07*  —  —  —  —  Cognitive dimension   Constant  3.63719    3.63811    3.75177    3.83827     Age  .00113  .03  .00072  .02  −.00143  −.04  .00072  .02   Age2  −.00033  −.21**  −.00021  −.10**  —  —  −.00015  −.07**   Female (1 = yes)  .00475  .00  .04130  .04  .01388  .02  .06021  .07**   Age * female  —  —  .00477  .09**  —  —  —  —  Reflective dimension   Constant  3.61939    3.70161    3.72336    3.85608     Age  .00333  .09  .00537  .14**  .00411  .10**  .00211  .06*   Age2  −.00018  −.11*  −.00036  −.17**  —  —  −.00027  −.13**   Age3  —  —  .000007  .12*  —  —  .000005  .09*   Female (1 = yes)  .00332  .00  −.05057  −.05*  −.03534  −.04  −.03628  −.04*  Compassionate dimension   Constant  3.38822    3.28081    3.27204    3.32472     Age  −.00014  −.00  .00067  .02  −.00049  −.01  −.00210  −.06**   Age2  −.00021  –.14**  —  —  .00027  .12**  .00012  .06**   Age3  —  —  —  —  −.000005  −.10*  —  —   Female (1 = yes)  .00437  .00  .11296  .13**  .10269  .11**  .09092  .10**  N  1,050  2,736  4,199  6,263  Note. b = unstandardized coefficient; β = standardized coefficient; Age and Education were centered at the mean. * p < .01. ** p ≤ .001. View Large Hypothesis 5 that the zenith of the curvilinear relation of age on wisdom and the cognitive wisdom dimension would be higher and occur later the higher the education or that age would be positively related to wisdom and the cognitive dimension for those with a higher education was not supported. As shown in Figure 2a, the association between age and wisdom was not significant for respondents with a high school or university education, but it followed an inverse U-curve for participants with a lower education. For those with an elementary education, the zenith was at age 39. For participants with an intermediate education, the interaction effect between age and gender on wisdom was significant, which resulted in a wisdom peak for men at age 45 and women at age 61 and a more gradual decline than for those with an elementary education. The association between age and the cognitive wisdom dimension was also not significant for respondents with a high school education and followed an inverse U-curve for the remaining three education groups with a significant interaction effect between age and gender for the intermediate education group again (see Figure 2b). Scores on the cognitive wisdom dimension tended to increase for respondents with an elementary education, men with an intermediate education, and those with a university education until early midlife (age 38–39) and for women with an intermediate education until age 50 and then decrease again, with a steeper decrease for the elementary education group than the other groups. However, up to the age of 72, average scores on the cognitive wisdom dimension tended to be highest for those with a university education. Figure 2. View largeDownload slide (a) Relation between age and three-dimensional wisdom by education, (b) relation between age and cognitive wisdom dimension by education. Legend for Predicted 3D-wisdom and the cognitive wisdom dimension: “............” represents elementary education. “- - - - - -” represents intermediate education–men. “-. -. -.” represents intermediate education–women. “__ __ __ _” represents high school education. “________”represents university education. (c) Relation between age and reflective wisdom dimension by education. (d) Relation between age and compassionate wisdom dimension by education. Legend for Predicted Reflective and Compassionate Wisdom Dimensions: “............” represents elementary education. “- - - - - -” represents intermediate education. “__ __ __ _” represents high school education. “________” represents university education. Figure 2. View largeDownload slide (a) Relation between age and three-dimensional wisdom by education, (b) relation between age and cognitive wisdom dimension by education. Legend for Predicted 3D-wisdom and the cognitive wisdom dimension: “............” represents elementary education. “- - - - - -” represents intermediate education–men. “-. -. -.” represents intermediate education–women. “__ __ __ _” represents high school education. “________”represents university education. (c) Relation between age and reflective wisdom dimension by education. (d) Relation between age and compassionate wisdom dimension by education. Legend for Predicted Reflective and Compassionate Wisdom Dimensions: “............” represents elementary education. “- - - - - -” represents intermediate education. “__ __ __ _” represents high school education. “________” represents university education. For participants with an elementary education, the relation between age and the reflective wisdom dimension followed an inverse U-curve with the zenith at age 46. For those with a high school education, age had a linear positive effect on the reflective wisdom dimension. However, a cubic association between age and the reflective wisdom dimension existed for those with an intermediate and university education, with average reflective scores increasing up to the age of 48 and 41, respectively, then staying relatively flat for the intermediate education group until the age of 60 and decreasing slightly for the university education group until the age of 69, before they began to rise again (see Figure 2c). The association between age and the compassionate wisdom dimension followed an inverse U-curve with the zenith at age 36 for the elementary education group, was not significant for the intermediate education group, and was U-shaped for the university education group with the nadir at age 45. A cubic relation existed between age and the compassionate dimension for participants with a high school education, with the nadir at age 37 and the zenith at age 70 (see Figure 2d). Because only 130 participants were over the age of 70 (with about 20% of cases each in the three lower educational groups and 40% with a college education), it was not clear how reliable the findings were for ages above 70. Therefore, we repeated the analyses with a cutoff age of 70, which resulted in almost identical outcomes for wisdom and the cognitive dimension but slightly different results for the other two wisdom dimensions. The relation between age and the reflective wisdom dimension became nonsignificant for participants with an elementary education, inverse U-shaped with the zenith at age 51 for those with an intermediate education, stayed linear and positive for those with a high school education, and retained its cubic form for those with a university education with an increase in average scores until age 35 followed by a slight decrease until age 58 and a subsequent increase in average scores. The association between age and the compassionate dimension of wisdom became nonsignificant for respondents with an elementary education, stayed nonsignificant for those with an intermediate education, became U-shaped for those with a high school education with the nadir at age 41, and stayed U-shaped for those with a university education with the nadir at age 45. Hence, Hypothesis 6 that increases in the reflective wisdom dimension in old age and in the compassionate wisdom dimension after midlife would only be observed for the two higher educational groups was supported only when the analysis was restricted to respondents up to the age of 70. Discussion Based on the bivariate correlations, age was unrelated to three-dimensional wisdom, negatively related to the cognitive and compassionate wisdom dimensions, and positively related to the reflective wisdom dimension. However, the bivariate correlations masked the curvilinear nature of the association between age and wisdom, particularly its individual dimensions, and variations by education and, to a lesser degree, gender. Although the relation between age and wisdom followed an inverse U-curve with the apex at midlife, replicating prior studies (Bergsma & Ardelt, 2012; Thomas et al., 2017; Webster et al., 2014) and supporting Hypothesis 1, the relatively flat curve was the result of almost opposing shapes for the cognitive and compassionate wisdom dimensions and an intermediate shape for the reflective dimension. Whereas Webster and colleagues (2014) found that the relation between age and wisdom, assessed by the SAWS, followed an inverse U-curve for the composite wisdom measure and four of its subscales (with the exception of reminiscence), the shapes for the three wisdom dimensions all varied in the present study. As predicted in Hypothesis 2 and the “fluid intelligence” view (Sternberg, 2005), average scores on the cognitive wisdom dimension were relatively stable from age 18 until midlife and then declined, which might have been due to decreases in fluid intelligence (Kievit et al., 2016), increases in rigidity or dogmatism (Meacham, 1990; Schultz & Searleman, 2002), and a decrease in openness to experiences, which Webster and colleagues (2014) also reported. Average scores on the reflective wisdom dimension first increased until midlife, when individuals might have a greater need to perceive phenomena and events from different perspectives to understand and manage their environment effectively, stayed relatively stable between 40 and 70, and increased again after age 70. This corroborates Hypothesis 3 and lends credence to the possibility that older adults might be impelled to engage in life review and (self-)reflection to overcome despair and master Erikson’s (1963) psychosocial task of achieving ego integrity when death looms ever closer on the horizon. Although research has found that life review is not limited to older age groups, the purpose of life review and reflection differs across the life course with death preparation being particularly relevant in old age (Staudinger, 2001). In direct opposition to the cognitive wisdom dimension, average scores on the compassionate wisdom dimension decreased slightly until midlife and then increased again as expected in Hypothesis 4. This pattern fits with Erikson’s (1963) psychosocial developmental pathway toward generativity at midlife, when self-centered ambitions and impulses are replaced by a concern for the welfare of the next generation and an increase in compassion for others. Greater average scores on the reflective and compassionate wisdom dimensions with advanced age are also compatible with an increased need for emotional regulation in old age and consistent with prior research (Chopik et al., 2017; Neff & Pommier, 2013; Scheibe & Carstensen, 2010). Educational attainment was positively related to wisdom and the cognitive and reflective wisdom dimensions and affected the shape of the age and wisdom relationship, including its three dimensions. For participants with an elementary education and men with an intermediate education, average wisdom scores first increased until midlife and then declined again, with a sharper decline for those with an elementary education. Interestingly, for women with an intermediate education, wisdom tended to increase until around age 60 and then decline only slightly. For participants with a high school or university education, age was unrelated to wisdom (but not positively related as predicted in Hypothesis 5), indicating that a higher education might have preserved an openness to experiences and prevented mental rigidity and dogmatism and, therefore, a loss of wisdom during the later years of life. A very similar pattern was observed for the association between age and the cognitive wisdom dimension, except that for participants with a university education, the relation also became curvilinear, with average scores increasing until midlife followed by a gradual decline but still higher average scores than any other educational group until around age 70. Women with an intermediate education tended to have scores on wisdom and the cognitive wisdom dimension similar to men with an intermediate education and those with an elementary education up to the age of about 35 and then tended to become more similar to those with a higher education after the age of about 45. It is likely that this reflects cohort differences as older women in Germany were probably encouraged to pursue an intermediate rather than a high school or university education even if they had the academic aptitude to obtain a higher educational degree (Breen, Luijkx, Müller, & Pollak, 2010). Hypothesis 6 that the expected growth in the reflective wisdom dimension in old age and the compassionate wisdom dimension after midlife would occur only for participants with a high school or university education was supported, but only for ages up to 70. Hence, participants with a higher education might be more likely to engage in life review and reflection in old age and generativity in midlife than those with a lower education. Beyond the age of 70, however, reflective dimension scores tended to increase for those with an intermediate education, and compassionate dimension scores tended to decline for those with a high school education. Yet, because the number of respondents over the age of 70 was relatively small, it is not clear how reliable these results are. The major limitations of the study are the cross-sectional and nonrepresentative nature of the data. Participants tended to be more educated than the general population of Germany and older adults over the age of 70 were underrepresented. Another limitation is that participants were aware that they took a “wisdom test,” which might have biased their answers in a presumably “wiser” direction, although Figures 1 and 2 show that mean values of wisdom scores and its three dimensions were not excessively inflated and comparable to other studies that used the 3D-WS without informing participants that the questionnaire contained a wisdom scale (e.g., Ardelt, 2010; Ardelt & Jeste, In press; Glück et al., 2013; Zacher, McKenna, & Rooney, 2013). Still, to be confident of the validity of the results, the study needs to be replicated with a large representative sample without informing participants that the study attempts to measure wisdom. Yet, the association between age and wisdom can ultimately only be tested with longitudinal data for different cohorts as cohort effects might be partly responsible for the results. The older cohorts in the sample experienced WWII and its aftermath in Germany and might have been more concerned with survival than the pursuit of wisdom (Koltko-Rivera, 2006). Despite its limitations, this study is the first with a sample large enough to analyze the relation between age and three-dimensional wisdom separately by educational groups. Although decreases in wisdom have been found during later life, the research indicated that the compassionate and reflective dimensions of wisdom might increase after midlife or during old age, respectively, for those with a higher education, which might be related to a growing need for emotion regulation (Carstensen & Mikels, 2005; Charles & Carstensen, 2009; Scheibe & Carstensen, 2010) and the successful mastery of Erikson’s (1963) psychosocial crises of generativity versus stagnation at midlife and ego integrity versus despair in old age. According to Erikson’s stage model, the resolution of these crises is facilitated by the successful mastery of four psychosocial crises in childhood and two in adolescence and emerging adulthood. Hence, positive human development in early life might lay the groundwork for greater wisdom later in life. Conflict of Interest None reported. Acknowledgements An earlier version of this study was presented at the 2016 European Sociological Association Research Network on Ageing in Europe Conference, Frankfurt, Germany. We like to thank the GEO technical team for providing the data and Lisa Christiansen for helping to clean up the data. References Aiken, L. S., & West, S. G. ( 1991). Multiple regression: Testing and interpreting interactions . Thousand Oaks, CA: Sage. Ardelt, M. ( 2000). Intellectual versus wisdom-related knowledge: The case for a different kind of learning in the later years of life. Educational Gerontology: An International Journal of Research and Practice , 26, 771– 789. doi:10.1080/036012700300001421 Google Scholar CrossRef Search ADS   Ardelt, M. ( 2003). Empirical assessment of a three-dimensional wisdom scale. Research on Aging , 25, 275– 324. doi:10.1177/0164027503025003004 Google Scholar CrossRef Search ADS   Ardelt, M. ( 2010). Are older adults wiser than college students? A comparison of two age cohorts. Journal of Adult Development , 17, 193– 207. doi:10.1007/s10804-009-9088-5 Google Scholar CrossRef Search ADS   Ardelt, M., & Jeste, D. ( In press). Wisdom and hard times: The ameliorating effect of wisdom on the negative association between adverse life events and well-being. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences . doi:10.1093/geronb/gbw137 Baltes, P. B., & Smith, J. ( 2008). The fascination of wisdom: Its nature, ontogeny, and function. Perspectives on Psychological Science: A Journal of the Association for Psychological Science , 3, 56– 64. doi:10.1111/j.1745-6916.2008.00062.x Google Scholar CrossRef Search ADS PubMed  Baltes, P. B., & Staudinger, U. M. ( 2000). Wisdom. A metaheuristic (pragmatic) to orchestrate mind and virtue toward excellence. The American Psychologist , 55, 122– 136. doi:10.1037/0003-066X.55.1.122 Google Scholar CrossRef Search ADS PubMed  Bergsma, A., & Ardelt, M. ( 2012). Self-reported wisdom and happiness: An empirical investigation. Journal of Happiness Studies , 13, 481– 499. doi:10.1007/s10902-011-9275-5 Google Scholar CrossRef Search ADS   Blanchard-Fields, F., & Norris, L. ( 1995). The development of wisdom. In M. A. Kimble, S. H. McFadden, J. W., J. J. Ellor, & Seeber (Eds.), Aging, spirituality, and religion. A handbook  (pp. 102– 118). Minneapolis, MN: Fortress Press. Breen, R., Luijkx, R., Müller, W., & Pollak, R. ( 2010). Long-term trends in educational inequality in Europe: Class inequalities and gender differences. European Sociological Review , 26, 31– 48. doi:10.1093/esr/jcp001 Google Scholar CrossRef Search ADS   Carstensen, L. L., & Mikels, J. A. ( 2005). At the intersection of emotion and cognition: Aging and the positivity effect. Current Directions in Psychological Science , 14, 117– 121. doi:10.1111/j.0963-7214.2005.00348.x Google Scholar CrossRef Search ADS   Charles, S. T., & Carstensen, L. L. ( 2009). Social and emotional aging. Annual Review of Psychology , 61, 383– 409. doi:10.1146/annurev.psych.093008.100448 Google Scholar CrossRef Search ADS   Cheraghi, F., Kadivar, P., Ardelt, M., Asgari, A., & Farzad, V. ( 2015). Gender as a moderator of the relation between age cohort and three-dimensional wisdom in Iranian culture. International Journal of Aging and Human Development , 81, 3– 26. doi:10.1177/0091415015616394 Google Scholar CrossRef Search ADS PubMed  Chopik, W. J., O’Brien, E., & Konrath, S. H. ( 2017). Differences in empathic concern and perspective taking across 63 countries. Journal of Cross-Cultural Psychology , 48, 23– 38. doi:10.1177/0022022116673910 Google Scholar CrossRef Search ADS   Clayton, V. P., & Birren, J. E. ( 1980). The development of wisdom across the life-span: A reexamination of an ancient topic. In Baltes P. B. & Brim O. G. (Eds.), Life-span development and behavior  (Vol. 3, pp. 103– 135). New York, NY: Academic Press. Erikson, E. H. ( 1959). Identity and the life cycle . New York, NY: International Universities Press, Inc. Erikson, E. H. ( 1963). Childhood and society . New York, NY: Norton. Ferrari, M., Kahn, A., Benayon, M., & Nero, J. ( 2011). Phronesis, sophia, and hochma: Developing wisdom in Islam and Judaism. Research in Human Development , 8, 128– 148. doi:10.1080/15427609.2011.568869 Google Scholar CrossRef Search ADS   Fischer, A. ( 2015). Wisdom - the answer to all the questions really worth asking. International Journal of Humanities and Social Science , 5, 73– 83. Glück, J. ( 2016). Weisheit: Die 5 Prinzipien des gelingenden Lebens (Wisdom: The 5 principles of a succeeding life) . München, Germany: Kösel. Glück, J., König, S., Naschenweng, K., Redzanowski, U., Dorner, L., Straßer, I., & Wiedermann, W. ( 2013). How to measure wisdom: Content, reliability, and validity of five measures. Frontiers in Psychology , 4, 405. doi:10.3389/fpsyg.2013.00405 Google Scholar CrossRef Search ADS PubMed  Horn, J. L., & Cattell, R. B. ( 1966). Age differences in primary mental ability factors. Journal of Gerontology , 21, 210– 220. doi:10.1093/geronj/21.2.210 Google Scholar CrossRef Search ADS PubMed  Kaufman, A. S. ( 2001). WAIS-III IQs, Horn’s theory, and generational changes from young adulthood to old age. Intelligence , 29, 131– 167. doi:10.1016/s0160-2896(00)00046-5 Google Scholar CrossRef Search ADS   Kekes, J. ( 1983). Wisdom. American Philosophical Quarterly , 20, 277– 286. Kievit, R. A., Davis, S. W., Griffiths, J., Correia, M. M., Cam-Can, & Henson, R. N. ( 2016). A watershed model of individual differences in fluid intelligence. Neuropsychologia , 91, 186– 198. doi:10.1016/j.neuropsychologia.2016.08.008 Google Scholar CrossRef Search ADS PubMed  Kitchener, K. S., & Brenner, H. G. ( 1990). Wisdom and reflective judgment: Knowing in the face of uncertainty. In Sternberg R. J. (Ed.), Wisdom: Its nature, origins, and development  (pp. 212– 229). New York, NY: Cambridge University Press. Google Scholar CrossRef Search ADS   Kohn, M. L., & Schooler, C. ( 1983). Work and personality: An inquiry into the impact of social stratification . Norwood, NJ: Ablex. Kohn, M. L., Zaborowski, W., Janicka, K., Mach, B. W., Khmelko, V., Slomczynski, K. M., … Podobnik, B. ( 2000). Complexity of activities and personality under conditions of radical social change: A comparative analysis of Poland and Ukraine. Social Psychology Quarterly , 63, 187– 208. doi:10.2307/2695868 Google Scholar CrossRef Search ADS   Koltko-Rivera, M. E. ( 2006). Rediscovering the later version of Maslow’s hierarchy of needs: Self-transcendence and opportunities for theory, research, and unification. Review of General Psychology , 10, 302– 317. doi:10.1037/1089-2680.10.4.302 Google Scholar CrossRef Search ADS   Kramer, D. A. ( 1990). Conceptualizing wisdom: The primacy of affect-cognition relations. In Sternberg R. J. (Ed.), Wisdom: Its nature, origins, and development  (pp. 279– 313). New York, NY: Cambridge University Press. Google Scholar CrossRef Search ADS   Levenson, M. R., Jennings, P. A., Aldwin, C. M., & Shiraishi, R. W. ( 2005). Self-transcendence: Conceptualization and measurement. International Journal of Aging & Human Development , 60, 127– 143. doi:10.2190/XRXM-FYRA-7U0X-GRC0 Google Scholar CrossRef Search ADS PubMed  Meacham, J. A. ( 1990). The loss of wisdom. In Sternberg R. J. (Ed.), Wisdom: Its nature, origins, and development  (pp. 181– 211). New York, NY: Cambridge University Press. Google Scholar CrossRef Search ADS   Murphy, S. L., Xu, J., & Kochanek, K. D. ( 2013). Deaths: final data for 2010. National Vital Statistics Reports: From the Centers for Disease Control and Prevention, National Center for Health Statistics, National Vital Statistics System , 61, 1– 117. Google Scholar PubMed  Neff, K. D., & Pommier, E. ( 2013). The relationship between self-compassion and other-focused concern among college undergraduates, community adults, and practicing meditators. Self and Identity , 12, 160– 176. doi:10.1080/15298868. 2011.649546 Google Scholar CrossRef Search ADS   Pasupathi, M., & Staudinger, U. M. ( 2001). Do advanced moral reasoners also show wisdom? Linking moral reasoning and wisdom-related knowledge and judgement. International Journal of Behavioral Development , 25, 401– 415. doi:10.1080/016502501316934833 Google Scholar CrossRef Search ADS   Pasupathi, M., Staudinger, U. M., & Baltes, P. B. ( 2001). Seeds of wisdom: Adolescents’ knowledge and judgment about difficult life problems. Developmental Psychology , 37, 351– 361. Google Scholar CrossRef Search ADS PubMed  Richardson, M. J., & Pasupathi, M. ( 2005). Young and growing wiser: Wisdom during adolescence and young adulthood. In R. J. Sternberg & J. Jordan (Eds.), A handbook of wisdom. Psychological perspectives  (pp. 139– 159). New York, NY: Cambridge University Press. Google Scholar CrossRef Search ADS   Rokeach, M. ( 1954). The nature and meaning of dogmatism. Psychological Review , 61, 194– 204. doi:10.1037/h0060752 Google Scholar CrossRef Search ADS PubMed  Scheibe, S., & Carstensen, L. L. ( 2010). Emotional aging: Recent findings and future trends. The Journals of Gerontology: Series B: Psychological Sciences and Social Sciences , 65, 135– 144. doi:10.1093/geronb/gbp132 Google Scholar CrossRef Search ADS   Schultz, P. W., & Searleman, A. ( 2002). Rigidity of thought and behavior: 100 years of research. Genetic, Social, and General Psychology Monographs , 128, 165– 207. Google Scholar PubMed  Staudinger, U. M. ( 2001). Life reflection: A social-cognitive analysis of life review. Review of General Psychology , 5, 148– 160. doi:10.1037/1089-2680.5.2.148 Google Scholar CrossRef Search ADS   Staudinger, U. M., & Glück, J. ( 2011). Psychological wisdom research: Commonalities and differences in a growing field. Annual Review of Psychology , 62, 215– 241. doi:10.1146/annurev.psych.121208.131659 Google Scholar CrossRef Search ADS PubMed  Staudinger, U. M., & Kunzmann, U. ( 2005). Positive adult personality development: Adjustment and/or growth? European Psychologist , 10, 320– 329. doi:10.1027/1016-9040.10.4.320 Google Scholar CrossRef Search ADS   Sternberg, R. J. ( 2005). Older but not wiser? The relationship between age and wisdom. Ageing International , 30, 5– 26. doi:10.1007/BF02681005 Google Scholar CrossRef Search ADS   Takahashi, M., & Overton, W. F. ( 2002). Wisdom: A culturally inclusive developmental perspective. International Journal of Behavioral Development , 26, 269– 277. doi:10.1080/01650250143000139 Google Scholar CrossRef Search ADS   Thomas, M. L., Bangen, K. J., Ardelt, M., & Jeste, D. V. ( 2017). Development of a 12-item abbreviated Three-Dimensional Wisdom Scale (3D-WS-12): Item selection and psychometric properties. Assessment , 24, 71– 82. doi:10.1177/1073191115595714 Google Scholar CrossRef Search ADS PubMed  Walsh, R. ( 2015). What is wisdom? Cross-cultural and cross-disciplinary syntheses. Review of General Psychology , 19, 278– 293. doi:10.1037/gpr0000045 Google Scholar CrossRef Search ADS   Webster, J. D. ( 2007). Measuring the character strength of wisdom. International Journal of Aging and Human Development , 65, 163– 183. doi:10.2190/AG.65.2.d Google Scholar CrossRef Search ADS PubMed  Webster, J. D., Westerhof, G. J., & Bohlmeijer, E. T. ( 2014). Wisdom and mental health across the lifespan. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 69, 209– 218. doi:10.1093/geronb/gbs121 Google Scholar CrossRef Search ADS PubMed  Wink, P., & Helson, R. ( 1997). Practical and transcendent wisdom: Their nature and some longitudinal findings. Journal of Adult Development , 4, 1– 15. doi:10.1007/BF02511845 Google Scholar CrossRef Search ADS   Zacher, H., McKenna, B., & Rooney, D. ( 2013). Effects of self-reported wisdom on happiness: Not much more than emotional intelligence? Journal of Happiness Studies , 14, 1697– 1716. doi:10.1007/s10902-012-9404-9 Google Scholar CrossRef Search ADS   © The Author(s) 2018. 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 Journals of Gerontology Series B: Psychological Sciences and Social Sciences Oxford University Press

The Relation Between Age and Three-Dimensional Wisdom: Variations by Wisdom Dimensions and Education

Loading next page...
 
/lp/ou_press/the-relation-between-age-and-three-dimensional-wisdom-variations-by-0Ph7IJg51Z
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. 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
1079-5014
eISSN
1758-5368
D.O.I.
10.1093/geronb/gbx182
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives Curvilinear relations between age and wisdom have been found in prior research with its peak either in young adulthood or in midlife. This study tested whether the association between age and three-dimensional wisdom differed for the cognitive, reflective, and compassionate (affective) wisdom dimensions and whether the results varied by education. Method OLS regression was used to analyze an online sample of 14,248 adults between the ages of 18 and 98 years (mean [M] = 36.46, standard deviation [SD] = 12.68) from four educational groups. Results The relation between age and three-dimensional wisdom was curvilinear and varied in shape by wisdom dimensions and education. The association between age and wisdom followed an inverse U-curve with the peak at midlife and almost opposing shapes for the cognitive and compassionate dimensions and an intermediate shape for the reflective dimension. Education was positively related to wisdom and affected the shape of the relationships. Discussion Because the association between age and wisdom appears to be curvilinear and varies by education, studies that test a linear association without separating the sample by education might arrive at the wrong conclusions. In general, education seems to promote the attainment of wisdom and also inoculate individuals against a decline in wisdom later in life. Successful aging, Human development, Gender It is often assumed colloquially that wisdom comes with age and experience, yet empirically and anecdotally this is not necessarily the case (Glück, 2016). Although life expectancy has increased during the last century (Murphy, Xu, & Kochanek, 2013), it appears that wisdom has not. In fact, wisdom seems to be rare among any age group, including the old. Life experiences might be a necessary but not sufficient condition for the development of wisdom. Instead, most experts and lay people agree that learning from life experiences is a prerequisite for the emergence of wisdom (Staudinger & Glück, 2011). Hence, if wisdom depends on the willingness to learn from experiences it might indeed increase with age but only up to a point until people believe that they have learned enough about life. If people become dogmatic in their views, think that they know more than they do, and do not realize the limits and uncertainty of knowledge (Kitchener & Brenner, 1990), wisdom might actually decrease with age. This might explain why a curvilinear association in the form of an inverse U-curve has been found between age and wisdom with the apex in middle age (Bergsma & Ardelt, 2012; Thomas, Bangen, Ardelt, & Jeste, 2017; Webster, Westerhof, & Bohlmeijer, 2014). Moreover, wisdom is a multifaceted concept that consists of cognitive, reflective, and affective elements (Ardelt, 2000; Blanchard-Fields & Norris, 1995; Kramer, 1990; Takahashi & Overton, 2002). Although the cognitive or knowledge aspect of wisdom might peak and then decline after a certain age because of an increase in rigidity or dogmatism (Schultz & Searleman, 2002), the reflective and emotional elements might not decrease but might even increase with age due to psychosocial growth (Erikson, 1963) and increases in perspective-taking skills, emotional regulation, and empathic concern or compassion for others (Chopik, O’Brien, & Konrath, 2017; Neff & Pommier, 2013; Scheibe & Carstensen, 2010). Individuals with a higher education are longer exposed to learning and might therefore also be more open and motivated to learn from life and develop greater wisdom. Hence, in addition to a positive association between education and various wisdom measures (Ardelt, 2003; Bergsma & Ardelt, 2012; Cheraghi, Kadivar, Ardelt, Asgari, & Farzad, 2015; Glück et al., 2013; Webster et al., 2014), the relation between age and wisdom and its individual elements might also depend on education. It is possible that the peak in the curvilinear association between age and wisdom comes earlier for those with a lower education or that wisdom does not decline with age for those with a higher education. The purpose of the current study was to examine whether the association between age and wisdom varied for the cognitive, reflective, and compassionate (affective) dimensions of the three-dimensional wisdom scale (Ardelt, 2003) and for different education levels, using a large German internet survey. The Concept of Wisdom Many definitions of wisdom exist (Staudinger & Glück, 2011). For example, wisdom has been defined as expert knowledge in the fundamental pragmatics of life (Baltes & Staudinger, 2000), orienting knowledge about what is good and right to live well (Fischer, 2015), “deep accurate insight and understanding of oneself and the central existential issues of life, plus skillful benevolent responsiveness” (Walsh, 2015, p. 282), and self-transcendence (Levenson, Jennings, Aldwin, & Shiraishi, 2005). We define and operationalize wisdom as the integration of cognitive, reflective, and compassionate personality characteristics (Ardelt, 2003) based on Clayton and Birren’s (1980) earlier research. The cognitive dimension describes a wise person’s desire to know a deeper truth and the ability to thoroughly understand the intrapersonal and interpersonal aspects of life, which includes an understanding and acceptance of the ultimate limits of knowledge, the positive and negative aspects of human nature, and life’s unpredictability and uncertainties. The reflective dimension refers to a wise person’s ability to perceive phenomena and events from multiple perspectives and to avoid subjectivity and projections, which requires self-examination, self-awareness, and self-insight and a reduction in self-centeredness. Diminished self-centeredness, in turn, is likely to result in greater understanding of life in general and other people in particular, leading to greater tolerance and sympathetic and compassionate love for others, which represents the compassionate dimension of wisdom. Although wisdom is defined as the integration of the three dimensions, greater insight into the overall association between age and wisdom might be gained by analyzing the relation between age and the three dimensions separately. Age and Wisdom Sternberg (2005) discussed five theoretical views of the age and wisdom relationship. According to the “received” view, which is often shared by laypeople, wisdom comes with age. That is, wisdom is assumed to be low or nonexistent earlier in life until it is received in old age. The “fluid intelligence” view suggests that wisdom is similar to fluid intelligence, or the “ability to think flexibly in novel ways” (Sternberg, 2005, p. 13) and, therefore, tends to increase until early adulthood, then levels off, until it starts to decline in middle age (Kievit, Davis, Griffiths, Correia, & Henson, 2016). The “crystallized intelligence” view describes wisdom as similar to crystallized intelligence or the ability to use acquired knowledge, experience, and skills effectively. Based on this view, wisdom increases continuously up to old age until it starts to decline, on average, about 10 years before one’s death due to the onset of disease, illness, and terminal decline (Kaufman, 2001). A fourth view combines the “fluid and crystallized intelligence” view. Wisdom is theorized to increase early in life until middle age when crystallized intelligence continues to rise while fluid intelligence begins to decline. This decrease in fluid intelligence results in a net negative effect on wisdom, so that wisdom increases until late middle age and then declines. Classical empirical work by Horn and Cattell (1966) showed that aspects of fluid intelligence such as associative memory, intellectual speed, and intellectual level, tended to decline considerably after early adulthood. By contrast, aspects of crystallized intelligence, including mechanical knowledge, ideational and associational fluency, and verbal comprehension, tended to increase with age but leveled off by midlife. These results have been replicated in later research with performance (fluid) intelligence and verbal (crystallized) intelligence (e.g., Kaufman, 2001). Hence, marked decreases in fluid intelligence and smaller gains in crystallized abilities by midlife predict an overall downward trend starting in late midlife for the “fluid and crystallized intelligence” view. Finally, the “declining wisdom” view advocated by Meacham (1990) states that wisdom declines with age, starting in early life. According to this view, children are considered wise and open to the world, but hardships and successes throughout life tend to reduce this wisdom. Hardships might diminish wisdom if they result in doubt and a loss of meaning and confidence in the world, whereas personal successes might decrease wisdom if they lead to a self-centered certainty and over-confidence in the world accompanied by dogmatism, prejudice, and intolerance. Three of the five views assume that wisdom increases during the early years of life due to normative personality and intellectual development and the exposure to education (Richardson & Pasupathi, 2005), which has been supported empirically (Ferrari, Kahn, Benayon, & Nero, 2011; Pasupathi, Staudinger, & Baltes, 2001). However, after an initial maturationah age in adulthood. Yet, older adults have a higher probability to be wiser than younger individuals because growth in wisdom is assumed to take time (Kekes, 1983). This means that wisdom might increase with age only for those individuals who actively pursue its acquisition and are able to learn from life experiences (Staudinger & Glück, 2011 increase in wisdom from childhood to adolescence and young adulthood, wisdom might not continue to develop witef>). Empirical evidence has shown that in adulthood, age is not necessarily linearly correlated with wisdom (Baltes & Staudinger, 2000; Glück et al., 2013; Levenson et al., 2005; Webster, 2007), although several studies have found a curvilinear association between age and wisdom, assessed by the Three-Dimensional Wisdom scale (3D-WS; Ardelt, 2003), integrating cognitive, reflective, and compassionate dimensions, and the Self-Assessed Wisdom Scale (SAWS; Webster, 2007), which is the average of critical life experiences, reminiscence/reflectiveness, openness to experience, emotional regulation, and humor, with an increase in wisdom until midlife followed by a decrease (Bergsma & Ardelt, 2012; Thomas et al., 2017; Webster et al., 2014). However, not all dimensions of wisdom might decline with age. While cognitive aspects of wisdom might follow the “fluid intelligence” view with an increase in wisdom until midlife and a decrease thereafter, noncognitive elements of wisdom might actually increase with age. For example, according to socioemotional selectivity theory (Carstensen & Mikels, 2005; Charles & Carstensen, 2009; Scheibe & Carstensen, 2010), older adults might be less interested in gaining greater knowledge but more likely than younger adults to engage in emotional regulation to maximize positive affect and minimize negative affect in an effort to make the best of their remaining time. One way to diminish negative affect is to perceive the larger picture, focus on the positive aspects, and comprehend others’ perspectives, which tends to decrease self-centeredness and increase compassion for others. Indeed, research has shown that age is positively related to emotional regulation, perspective-taking skills, and empathic concern or compassion for others (Chopik et al., 2017; Neff & Pommier, 2013; Scheibe & Carstensen, 2010). Research also has found that a motivation for psychosocial growth and favorable psychological, social, and historical conditions might foster the development of wisdom (Baltes & Smith, 2008). For example, among German adults between the ages of 20 and 87, age was positively related to wisdom-related knowledge for those with higher levels of moral reasoning but not for those with moral reasoning scores below the median (Pasupathi & Staudinger, 2001). In an Iranian sample of adults between 18 and 80 years of age, age was positively correlated with three-dimensional wisdom and the reflective and compassionate wisdom dimensions for those with a higher education but unrelated for those with a lower education. Yet, age and the cognitive wisdom dimension were unrelated among the higher educated participants and were negatively correlated among those with a lower education (Cheraghi et al., 2015). Moreover, in a study comparing college students (mean [M] = 21 years) with older adults between ages 52 and 87 (M = 71 years), college-educated older adults had higher average three-dimensional wisdom scores and scores on the reflective and compassionate wisdom dimensions than young college students, but older adults without a college degree had significantly lower average wisdom scores and scores on the cognitive wisdom dimension than college-educated older adults and college students (Ardelt, 2010). In longitudinal research, Wink and Helson (1997) found that highly educated participants, and clinical psychologists in particular, experienced gains in practical wisdom from age 27 to 52. Taken together, these studies support the “crystallized intelligence” view for the noncognitive dimensions of wisdom for people who are highly educated and/or have the motivation to grow in wisdom. Education might provide the “luxury” to continue to be curious about the world, one’s place in it, and the deeper meaning of life. Adults with a lower education and employment that is characterized by physical labor and/or repetitive work might be too exhausted to engage in activities of psychosocial growth. Research by Kohn and colleagues (Kohn & Schooler, 1983; Kohn et al., 2000) demonstrated that individuals with a higher education were more likely to endorse autonomy, self-directedness, and openness to change, which was reflected and reinforced in their work environment, whereas those with a lower education tended to hold authoritarian, conformative beliefs and were employed in less complex and more supervised, standardized jobs. This resembles Staudinger and Kunzmann’s (2005) distinction between personality growth and personality adjustment. Wisdom is an indicator of personality growth, which requires openness to experiences and a questioning attitude toward societal norms rather than conforming and adjusting to the prevailing societal standards. Therefore, a higher education facilitates personality growth, while a lower education encourages personality adjustment, particularly through work requirements. Hence, a positive association between age and wisdom might be more likely among highly educated individuals than among individuals with a lower education. Yet, those differences and especially nonlinear differences by educational attainment might not be visible in small samples with participants from diverse educational backgrounds. The Present Study The goal of the present study was twofold. First, we examined how age was related to three-dimensional wisdom in adulthood and whether this relation differed for the three individual dimensions. Second, we analyzed whether the associations between age and wisdom and the three individual dimensions varied by educational groups. While the five theoretical views of the age and wisdom relationship assume a unidimensional conception of wisdom, we predicted that the associations between age and the three individual wisdom dimensions would not follow the same pattern. We hypothesized that the relation between age and composite three-dimensional wisdom in adulthood followed an inverse U-curve with the peak at midlife (Hypothesis 1), based on prior studies (Bergsma & Ardelt, 2012; Thomas et al., 2017; Webster et al., 2014) and similar to the “fluid and crystallized intelligence” view (Sternberg, 2005). Webster and colleagues (2014, p. 215) found that older adults tend to have more life experiences but also less openness to new ideas, whereas younger adults tend to be less experienced but more accepting of new knowledge, concluding that “midlife adults have a critical mass of life experiences and the cognitive strengths and emotional resources to process such events.” Yet, people at the top of their careers and/or having raised their children to become independent adults might feel that they know how the world works rather than understanding, as Socrates did, that there is much more to know. If individuals do not continue to learn from experiences, they might become dogmatic and unwilling to search for a deeper truth or to engage in reflective thinking and self-reflection to question their existing beliefs (Meacham, 1990; Rokeach, 1954). This curvilinear association might be most pronounced for the cognitive wisdom dimension, which represents a desire to know a deeper truth, with a steeper decline after midlife (Hypothesis 2). The reflective wisdom dimension, which consists of perspective-taking, self-reflection, and the avoidance of blame, might first follow the same shape but might actually increase in old age when individuals use emotional regulation to maximize positive affect (Carstensen & Mikels, 2005; Charles & Carstensen, 2009; Scheibe & Carstensen, 2010) and struggle with the psychosocial crisis of integrity versus despair (Hypothesis 3). According to Erikson’s (1963) stage model of human development, older people can achieve ego integrity and greater wisdom when they engage in life review and reflect on the past as it relates to the present to learn acceptance of the totality of their lives without despairing over past failures or mistakes, missed opportunities, or the nearing of death (Staudinger, 2001). Later life might also provide more time for this kind of self-reflection, particularly after retirement (Ardelt, 2000). The compassionate wisdom dimension requires a reduction in self-centeredness to develop sympathy and compassion for others. This reduction in self-centeredness might begin in midlife and continue into old age if individuals master Erikson’s psychosocial crisis of generativity versus self-absorption or stagnation. During midlife, many adults display an “interest in establishing and guiding the next generation” (Erikson, 1959, p. 97) by focusing more on the well-being of others and contributing to the success and well-being of future generations, not just their own children or grandchildren. Hence, the compassionate wisdom dimension was expected to increase, starting in midlife (Hypothesis 4). Individuals with a higher education are probably more likely to continue to learn new things and, therefore, might be more open and motivated to learn from life experiences than those with a lower education. A higher education is also more likely to result in leadership positions, which makes it easier to engage in generativity. Based on this reasoning, we predicted that the zenith of the curvilinear relation of age on wisdom and the cognitive wisdom dimension would be higher and occur later for individuals with a higher education than for those with a lower education or that age might even be positively related to wisdom and the cognitive dimension for those with a higher education (Hypothesis 5). Similarly, we hypothesized that individuals with a higher education might be more interested in life review and reflection in old age and motivated to engage in generativity in midlife than those with a lower education with increases in the reflective wisdom dimension during later life and in the compassionate wisdom dimension after midlife only expected for the high school and university education groups (Hypothesis 6). In exploratory analyses, we also tested for possible moderation effects of gender on the association between age and/or education on wisdom and its three dimensions. Method Procedure In 2006, the German magazine GEO (equivalent to National Geographic in the United States) published a three-part series about wisdom in their March, April, and May issues and offered their readers the opportunity to take an online “wisdom test.” Participants received an average score for the three-dimensional wisdom scale and its cognitive, reflective, and compassionate dimensions. General interpretations of scores below the midpoint of the scales (a score of <3), slightly above the midpoint (a score between 3 and <4), and in the top of the scale (a score between 4 and 5) were provided by the author of the GEO wisdom article after feedback from the senior author. The interpretations included general statements and tips about self-improvement and emphasized the positive even for those who scored below the midpoint. In addition to wisdom, age, gender, and education were assessed. Readers were informed that the anonymous data would be analyzed in a study. GEO technicians collected the raw data that were submitted between March 17 and May 5, 2006, including time and date of submission. Measures Three-dimensional wisdom was assessed by the cognitive, reflective, and compassionate (affective) dimensions of the 3D-WS (Ardelt, 2003) on two 5-point scales (1 = strongly agree and 5 = strongly disagree or 1 = definitely true of myself and 5 = not true of myself). The 14 items of the cognitive wisdom dimension measured the ability or willingness to understand a situation or phenomenon thoroughly, knowledge of the positive and negative aspects of human nature, an acknowledgement of ambiguity and uncertainty in life, and the ability to make important decision despite life’s unpredictability and uncertainties (e.g., Ignorance is bliss—reversed). The 12 items of the reflective wisdom dimension gauged the ability and willingness to look at phenomena and events from different perspectives and the absence of bitterness, subjectivity, and projections (e.g., I always try to look at all sides of a problem). The 13 items of the compassionate wisdom dimension assessed the presence of positive, caring, and nurturing emotions and behavior and the absence of indifferent or negative emotions and behavior toward others (e.g., Sometimes I feel a real compassion for everyone). All items were scored in the direction of greater wisdom before calculating the average for each wisdom dimension. Cronbach’s α was .72, .77, and .68 for the cognitive, reflective, and compassionate dimensions, respectively, and .74 for the composite wisdom scale consisting of the average of the three dimensions rather than the 39 items (Cronbach’s α was .85 for the 39 items). The cognitive dimension correlated r = .51 with the reflective dimension and r = .45 with the compassionate dimension, and the correlation between the reflective and compassionate wisdom dimensions was r = .48 (all p < .001). Age was measured in years and gender as 0 = male and 1 = female. Education had four categories (1 = elementary education, 2 = intermediate education, 3 = high school education, 4 = university education). To test for nonlinear associations between age and wisdom and all combinations of interaction effects between age, education, and gender on wisdom, age and education were centered at the mean to reduce multicollinearity (Aiken & West, 1991). Sample The original data set contained 28,538 cases. Unfortunately, the GEO technicians did not block multiple submissions from the same IP address nor did they collect the IP addresses of participants, which required a manual clean-up of the data. Initially, we deleted cases with the same demographics, the same score on all wisdom measures, and the same timestamp (to the minute), as these participants likely hit the submit button more than once. All cases that matched exactly on each wisdom item, differed by only one demographic variable, and were entered within 10 minutes of each other were deleted, because we believe curious respondents were trying to determine how changes in the demographic variables influenced the final wisdom score. These procedures led to the removal of 13,309 cases. Subsequently, all cases with missing values on gender, education, and wisdom were removed (234 cases). We also performed an internal validity check on the 3D-WS and deleted 46 cases that strongly endorsed items or strongly disagreed with items that contradicted each other (e.g., strongly endorsing the item “If I see people in need, I try to help them one way or another” and also the items “It’s not really my problem if others are in trouble and need help” and “I often have not comforted another when he or she needed it”). Twenty-one outliers with extremely low or “perfect” 3D-WS scores were also deleted, because they did not appear to reflect valid entries. Finally, we removed 701 participants under the age of 18 or with an age of 99. We were concerned that the nine people who entered 99 did not want to reveal their real age. Minors were deleted for ethical reasons as we did not ask their parents’ permission to participate in the research and because some of the minor ages were clearly fake (e.g., ages of 0–5). After cleaning the data, 14,248 respondents between the ages of 18 and 98 (M = 36.46, standard deviation [SD] = 12.68, median = 35) remained. Approximately half (48.5%) of the participants were women and 7.4% reported having an elementary education (Haupschule), 19.2% an intermediate education (Realschule), 29.5% a high school education (Gymnasium), and 44.0% a university education. The percentage who earn a high school degree (Abitur) in Germany is much lower than in the United States and was only 33% in 2006, compared to 28% who received an elementary education and 39% who obtained an intermediate education (http://www.deutschlandinzahlen.de/tab/deutschland/bildung/schule/allgemeinbildende/schulabsolventen-insgesamt). Given these statistics and considering that not everyone with a high school degree will get a university education, it is clear that the present sample is biased in the direction of a higher education. Results Bivariate Correlations In bivariate correlations, age was unrelated to the 3D-WS, but negatively correlated with the cognitive (r = −.05) and compassionate (r = −.04) wisdom dimensions and positively with the reflective wisdom dimension (r = .07). Education was positively associated with the 3D-WS (r = .16) and with the cognitive (r = .20), reflective (r = .16), and compassionate (r = .04) wisdom dimensions but negatively with age (r = −.09). Being female was positively correlated with the 3D-WS (r = .04), the cognitive (r = .05), and compassionate (r = .10) wisdom dimensions but negatively with the reflective wisdom dimension (r = −.04) and age (r = −.04). Women were less likely than men to report having an elementary education (6% vs. 9%) and a university education (42% vs. 45%) but more likely than men to indicate an intermediate education (21% vs. 18%). All of these correlations were significant at p < .001, although some of the effect sizes were quite small. Multivariate Regression Analyses for the Full Sample To perform the multivariate OLS regression analyses with the full sample, the significance level was set at p ≤.001 due to the large sample size. First, age, gender, and education were entered into the model. Age-square was added if it was significant and then age-cube if it was significant. We also tested whether any of the interaction effects with education and/or gender were statistically significant, using the FORWARD command in SPSS 24. The results for the full sample are displayed in Table 1 and in Figure 1 with the original age range. As predicted in Hypothesis 1, the association between age and wisdom followed an inverse U-curve with the peak at age 43. However, the curve was rather flat as Figure 1 shows. The curvilinear relation was more pronounced for the cognitive wisdom dimension, confirming Hypothesis 2. In fact, a cubic association existed between age and the cognitive wisdom dimension, but with a local minimum at age 17, beyond the current age range, and a maximum at age 42. Figure 1 shows that age tended to be unrelated to the cognitive wisdom dimension until it started to decline after age 42. There was also a cubic relation between age and the reflective wisdom dimension. As expected in Hypothesis 3, the reflective wisdom dimension tended to increase until midlife (age 46), then was relatively stable as depicted in Figure 1, and subsequently tended to increase again after the age of 71. Hypothesis 4 was also supported. The association between age and the compassionate wisdom dimension followed a quadratic U-curve with slight decreases, on average, in the compassionate wisdom dimension until midlife and increases after age 46. However, there was a significant interaction effect between age-squared and education, suggesting that the shape differed by educational group. Because education was centered at the mean (M = 3.10, SD = 0.96), the curvilinear association shown in Figure 1 represents respondents with a high school education the most. The curvilinear shape also applied to those with a university education but with the nadir at age 44. However, for respondents with an elementary education, the shape became an inverse U-curve with the peak at age 37, and for participants with an intermediate education, age appeared to be unrelated to the compassionate wisdom dimension. After controlling for age, education was positively related to the 3D-WS and the cognitive and reflective wisdom dimensions but not the compassionate wisdom dimension. Table 1. Relation Between Age and Three-Dimensional Wisdom Controlling for Education and Gender; OLS Regression Analyses   3D-wisdom  Cognitive dimension  Reflective dimension  Compassionate dimension  Independent variables  b  β  b  β  b  β  b  β  Constant  3.61439    3.75669    3.77487    3.31001    Age  .00093  .03**  .00109  .03  .00337  .09**  −.00179  −.05**  Age2  −.00007  −.04**  −.00001  −.03  −.00022  −.11**  .00009  .05**  Age3  —  —  −.000003  −.06**  .000003  .06**  —  —  Education  .06401  .16**  .09624  .20**  .07926  .16**  .00417  .01  Female (1 = yes)  .03248  .04**  .04024  .04**  −.03547  −.04**  .09173  .10**  Age * Education  —  —  —  —  —  —  −.00092  −.03  Age2 * Education  —  —  —  —  —  —  .00009  .06**  Adjusted R2  .03**  .05**  .04**  .02**    3D-wisdom  Cognitive dimension  Reflective dimension  Compassionate dimension  Independent variables  b  β  b  β  b  β  b  β  Constant  3.61439    3.75669    3.77487    3.31001    Age  .00093  .03**  .00109  .03  .00337  .09**  −.00179  −.05**  Age2  −.00007  −.04**  −.00001  −.03  −.00022  −.11**  .00009  .05**  Age3  —  —  −.000003  −.06**  .000003  .06**  —  —  Education  .06401  .16**  .09624  .20**  .07926  .16**  .00417  .01  Female (1 = yes)  .03248  .04**  .04024  .04**  −.03547  −.04**  .09173  .10**  Age * Education  —  —  —  —  —  —  −.00092  −.03  Age2 * Education  —  —  —  —  —  —  .00009  .06**  Adjusted R2  .03**  .05**  .04**  .02**  Note. N = 14,248; b = unstandardized coefficient; β = standardized coefficient; Age and Education were centered at the mean. ** p ≤ .001. View Large Table 1. Relation Between Age and Three-Dimensional Wisdom Controlling for Education and Gender; OLS Regression Analyses   3D-wisdom  Cognitive dimension  Reflective dimension  Compassionate dimension  Independent variables  b  β  b  β  b  β  b  β  Constant  3.61439    3.75669    3.77487    3.31001    Age  .00093  .03**  .00109  .03  .00337  .09**  −.00179  −.05**  Age2  −.00007  −.04**  −.00001  −.03  −.00022  −.11**  .00009  .05**  Age3  —  —  −.000003  −.06**  .000003  .06**  —  —  Education  .06401  .16**  .09624  .20**  .07926  .16**  .00417  .01  Female (1 = yes)  .03248  .04**  .04024  .04**  −.03547  −.04**  .09173  .10**  Age * Education  —  —  —  —  —  —  −.00092  −.03  Age2 * Education  —  —  —  —  —  —  .00009  .06**  Adjusted R2  .03**  .05**  .04**  .02**    3D-wisdom  Cognitive dimension  Reflective dimension  Compassionate dimension  Independent variables  b  β  b  β  b  β  b  β  Constant  3.61439    3.75669    3.77487    3.31001    Age  .00093  .03**  .00109  .03  .00337  .09**  −.00179  −.05**  Age2  −.00007  −.04**  −.00001  −.03  −.00022  −.11**  .00009  .05**  Age3  —  —  −.000003  −.06**  .000003  .06**  —  —  Education  .06401  .16**  .09624  .20**  .07926  .16**  .00417  .01  Female (1 = yes)  .03248  .04**  .04024  .04**  −.03547  −.04**  .09173  .10**  Age * Education  —  —  —  —  —  —  −.00092  −.03  Age2 * Education  —  —  —  —  —  —  .00009  .06**  Adjusted R2  .03**  .05**  .04**  .02**  Note. N = 14,248; b = unstandardized coefficient; β = standardized coefficient; Age and Education were centered at the mean. ** p ≤ .001. View Large Figure 1. View largeDownload slide Relation between age and three-dimensional wisdom by wisdom dimensions. “____” represents composite three-dimensional wisdom. “............” represents cognitive wisdom dimension. “- - - - - ” represents reflective wisdom dimension. “__ __ __ ” represents compassionate wisdom dimension. Figure 1. View largeDownload slide Relation between age and three-dimensional wisdom by wisdom dimensions. “____” represents composite three-dimensional wisdom. “............” represents cognitive wisdom dimension. “- - - - - ” represents reflective wisdom dimension. “__ __ __ ” represents compassionate wisdom dimension. Multivariate Regression Analyses for Separate Educational Groups The same procedure as for the full sample was followed for the separate analyses by educational group (see Table 2). Because the sample sizes were smaller, the significance level was increased to p <.01. Table 2. Relation between Age and Three-Dimensional Wisdom by Education Controlling for Gender; OLS Regression Analyses Education  Elementary  Intermediate  High school  University  Wisdom Measure  b  β  b  β  b  β  b  β  3D-Wisdom   Constant  3.54827    3.53064    3.59651    3.66183     Age  .00144  .05  .00178  .06  .00025  .01  −.00005  −.00   Age2  −.00024  −.18**  −.00010  −.06*  —  —  —  —   Female (1 = yes)  .00415  .01  .03196  .04  .02574  .03  .03864  .05**   Age * female  —  —  .00318  .07*  —  —  —  —  Cognitive dimension   Constant  3.63719    3.63811    3.75177    3.83827     Age  .00113  .03  .00072  .02  −.00143  −.04  .00072  .02   Age2  −.00033  −.21**  −.00021  −.10**  —  —  −.00015  −.07**   Female (1 = yes)  .00475  .00  .04130  .04  .01388  .02  .06021  .07**   Age * female  —  —  .00477  .09**  —  —  —  —  Reflective dimension   Constant  3.61939    3.70161    3.72336    3.85608     Age  .00333  .09  .00537  .14**  .00411  .10**  .00211  .06*   Age2  −.00018  −.11*  −.00036  −.17**  —  —  −.00027  −.13**   Age3  —  —  .000007  .12*  —  —  .000005  .09*   Female (1 = yes)  .00332  .00  −.05057  −.05*  −.03534  −.04  −.03628  −.04*  Compassionate dimension   Constant  3.38822    3.28081    3.27204    3.32472     Age  −.00014  −.00  .00067  .02  −.00049  −.01  −.00210  −.06**   Age2  −.00021  –.14**  —  —  .00027  .12**  .00012  .06**   Age3  —  —  —  —  −.000005  −.10*  —  —   Female (1 = yes)  .00437  .00  .11296  .13**  .10269  .11**  .09092  .10**  N  1,050  2,736  4,199  6,263  Education  Elementary  Intermediate  High school  University  Wisdom Measure  b  β  b  β  b  β  b  β  3D-Wisdom   Constant  3.54827    3.53064    3.59651    3.66183     Age  .00144  .05  .00178  .06  .00025  .01  −.00005  −.00   Age2  −.00024  −.18**  −.00010  −.06*  —  —  —  —   Female (1 = yes)  .00415  .01  .03196  .04  .02574  .03  .03864  .05**   Age * female  —  —  .00318  .07*  —  —  —  —  Cognitive dimension   Constant  3.63719    3.63811    3.75177    3.83827     Age  .00113  .03  .00072  .02  −.00143  −.04  .00072  .02   Age2  −.00033  −.21**  −.00021  −.10**  —  —  −.00015  −.07**   Female (1 = yes)  .00475  .00  .04130  .04  .01388  .02  .06021  .07**   Age * female  —  —  .00477  .09**  —  —  —  —  Reflective dimension   Constant  3.61939    3.70161    3.72336    3.85608     Age  .00333  .09  .00537  .14**  .00411  .10**  .00211  .06*   Age2  −.00018  −.11*  −.00036  −.17**  —  —  −.00027  −.13**   Age3  —  —  .000007  .12*  —  —  .000005  .09*   Female (1 = yes)  .00332  .00  −.05057  −.05*  −.03534  −.04  −.03628  −.04*  Compassionate dimension   Constant  3.38822    3.28081    3.27204    3.32472     Age  −.00014  −.00  .00067  .02  −.00049  −.01  −.00210  −.06**   Age2  −.00021  –.14**  —  —  .00027  .12**  .00012  .06**   Age3  —  —  —  —  −.000005  −.10*  —  —   Female (1 = yes)  .00437  .00  .11296  .13**  .10269  .11**  .09092  .10**  N  1,050  2,736  4,199  6,263  Note. b = unstandardized coefficient; β = standardized coefficient; Age and Education were centered at the mean. * p < .01. ** p ≤ .001. View Large Table 2. Relation between Age and Three-Dimensional Wisdom by Education Controlling for Gender; OLS Regression Analyses Education  Elementary  Intermediate  High school  University  Wisdom Measure  b  β  b  β  b  β  b  β  3D-Wisdom   Constant  3.54827    3.53064    3.59651    3.66183     Age  .00144  .05  .00178  .06  .00025  .01  −.00005  −.00   Age2  −.00024  −.18**  −.00010  −.06*  —  —  —  —   Female (1 = yes)  .00415  .01  .03196  .04  .02574  .03  .03864  .05**   Age * female  —  —  .00318  .07*  —  —  —  —  Cognitive dimension   Constant  3.63719    3.63811    3.75177    3.83827     Age  .00113  .03  .00072  .02  −.00143  −.04  .00072  .02   Age2  −.00033  −.21**  −.00021  −.10**  —  —  −.00015  −.07**   Female (1 = yes)  .00475  .00  .04130  .04  .01388  .02  .06021  .07**   Age * female  —  —  .00477  .09**  —  —  —  —  Reflective dimension   Constant  3.61939    3.70161    3.72336    3.85608     Age  .00333  .09  .00537  .14**  .00411  .10**  .00211  .06*   Age2  −.00018  −.11*  −.00036  −.17**  —  —  −.00027  −.13**   Age3  —  —  .000007  .12*  —  —  .000005  .09*   Female (1 = yes)  .00332  .00  −.05057  −.05*  −.03534  −.04  −.03628  −.04*  Compassionate dimension   Constant  3.38822    3.28081    3.27204    3.32472     Age  −.00014  −.00  .00067  .02  −.00049  −.01  −.00210  −.06**   Age2  −.00021  –.14**  —  —  .00027  .12**  .00012  .06**   Age3  —  —  —  —  −.000005  −.10*  —  —   Female (1 = yes)  .00437  .00  .11296  .13**  .10269  .11**  .09092  .10**  N  1,050  2,736  4,199  6,263  Education  Elementary  Intermediate  High school  University  Wisdom Measure  b  β  b  β  b  β  b  β  3D-Wisdom   Constant  3.54827    3.53064    3.59651    3.66183     Age  .00144  .05  .00178  .06  .00025  .01  −.00005  −.00   Age2  −.00024  −.18**  −.00010  −.06*  —  —  —  —   Female (1 = yes)  .00415  .01  .03196  .04  .02574  .03  .03864  .05**   Age * female  —  —  .00318  .07*  —  —  —  —  Cognitive dimension   Constant  3.63719    3.63811    3.75177    3.83827     Age  .00113  .03  .00072  .02  −.00143  −.04  .00072  .02   Age2  −.00033  −.21**  −.00021  −.10**  —  —  −.00015  −.07**   Female (1 = yes)  .00475  .00  .04130  .04  .01388  .02  .06021  .07**   Age * female  —  —  .00477  .09**  —  —  —  —  Reflective dimension   Constant  3.61939    3.70161    3.72336    3.85608     Age  .00333  .09  .00537  .14**  .00411  .10**  .00211  .06*   Age2  −.00018  −.11*  −.00036  −.17**  —  —  −.00027  −.13**   Age3  —  —  .000007  .12*  —  —  .000005  .09*   Female (1 = yes)  .00332  .00  −.05057  −.05*  −.03534  −.04  −.03628  −.04*  Compassionate dimension   Constant  3.38822    3.28081    3.27204    3.32472     Age  −.00014  −.00  .00067  .02  −.00049  −.01  −.00210  −.06**   Age2  −.00021  –.14**  —  —  .00027  .12**  .00012  .06**   Age3  —  —  —  —  −.000005  −.10*  —  —   Female (1 = yes)  .00437  .00  .11296  .13**  .10269  .11**  .09092  .10**  N  1,050  2,736  4,199  6,263  Note. b = unstandardized coefficient; β = standardized coefficient; Age and Education were centered at the mean. * p < .01. ** p ≤ .001. View Large Hypothesis 5 that the zenith of the curvilinear relation of age on wisdom and the cognitive wisdom dimension would be higher and occur later the higher the education or that age would be positively related to wisdom and the cognitive dimension for those with a higher education was not supported. As shown in Figure 2a, the association between age and wisdom was not significant for respondents with a high school or university education, but it followed an inverse U-curve for participants with a lower education. For those with an elementary education, the zenith was at age 39. For participants with an intermediate education, the interaction effect between age and gender on wisdom was significant, which resulted in a wisdom peak for men at age 45 and women at age 61 and a more gradual decline than for those with an elementary education. The association between age and the cognitive wisdom dimension was also not significant for respondents with a high school education and followed an inverse U-curve for the remaining three education groups with a significant interaction effect between age and gender for the intermediate education group again (see Figure 2b). Scores on the cognitive wisdom dimension tended to increase for respondents with an elementary education, men with an intermediate education, and those with a university education until early midlife (age 38–39) and for women with an intermediate education until age 50 and then decrease again, with a steeper decrease for the elementary education group than the other groups. However, up to the age of 72, average scores on the cognitive wisdom dimension tended to be highest for those with a university education. Figure 2. View largeDownload slide (a) Relation between age and three-dimensional wisdom by education, (b) relation between age and cognitive wisdom dimension by education. Legend for Predicted 3D-wisdom and the cognitive wisdom dimension: “............” represents elementary education. “- - - - - -” represents intermediate education–men. “-. -. -.” represents intermediate education–women. “__ __ __ _” represents high school education. “________”represents university education. (c) Relation between age and reflective wisdom dimension by education. (d) Relation between age and compassionate wisdom dimension by education. Legend for Predicted Reflective and Compassionate Wisdom Dimensions: “............” represents elementary education. “- - - - - -” represents intermediate education. “__ __ __ _” represents high school education. “________” represents university education. Figure 2. View largeDownload slide (a) Relation between age and three-dimensional wisdom by education, (b) relation between age and cognitive wisdom dimension by education. Legend for Predicted 3D-wisdom and the cognitive wisdom dimension: “............” represents elementary education. “- - - - - -” represents intermediate education–men. “-. -. -.” represents intermediate education–women. “__ __ __ _” represents high school education. “________”represents university education. (c) Relation between age and reflective wisdom dimension by education. (d) Relation between age and compassionate wisdom dimension by education. Legend for Predicted Reflective and Compassionate Wisdom Dimensions: “............” represents elementary education. “- - - - - -” represents intermediate education. “__ __ __ _” represents high school education. “________” represents university education. For participants with an elementary education, the relation between age and the reflective wisdom dimension followed an inverse U-curve with the zenith at age 46. For those with a high school education, age had a linear positive effect on the reflective wisdom dimension. However, a cubic association between age and the reflective wisdom dimension existed for those with an intermediate and university education, with average reflective scores increasing up to the age of 48 and 41, respectively, then staying relatively flat for the intermediate education group until the age of 60 and decreasing slightly for the university education group until the age of 69, before they began to rise again (see Figure 2c). The association between age and the compassionate wisdom dimension followed an inverse U-curve with the zenith at age 36 for the elementary education group, was not significant for the intermediate education group, and was U-shaped for the university education group with the nadir at age 45. A cubic relation existed between age and the compassionate dimension for participants with a high school education, with the nadir at age 37 and the zenith at age 70 (see Figure 2d). Because only 130 participants were over the age of 70 (with about 20% of cases each in the three lower educational groups and 40% with a college education), it was not clear how reliable the findings were for ages above 70. Therefore, we repeated the analyses with a cutoff age of 70, which resulted in almost identical outcomes for wisdom and the cognitive dimension but slightly different results for the other two wisdom dimensions. The relation between age and the reflective wisdom dimension became nonsignificant for participants with an elementary education, inverse U-shaped with the zenith at age 51 for those with an intermediate education, stayed linear and positive for those with a high school education, and retained its cubic form for those with a university education with an increase in average scores until age 35 followed by a slight decrease until age 58 and a subsequent increase in average scores. The association between age and the compassionate dimension of wisdom became nonsignificant for respondents with an elementary education, stayed nonsignificant for those with an intermediate education, became U-shaped for those with a high school education with the nadir at age 41, and stayed U-shaped for those with a university education with the nadir at age 45. Hence, Hypothesis 6 that increases in the reflective wisdom dimension in old age and in the compassionate wisdom dimension after midlife would only be observed for the two higher educational groups was supported only when the analysis was restricted to respondents up to the age of 70. Discussion Based on the bivariate correlations, age was unrelated to three-dimensional wisdom, negatively related to the cognitive and compassionate wisdom dimensions, and positively related to the reflective wisdom dimension. However, the bivariate correlations masked the curvilinear nature of the association between age and wisdom, particularly its individual dimensions, and variations by education and, to a lesser degree, gender. Although the relation between age and wisdom followed an inverse U-curve with the apex at midlife, replicating prior studies (Bergsma & Ardelt, 2012; Thomas et al., 2017; Webster et al., 2014) and supporting Hypothesis 1, the relatively flat curve was the result of almost opposing shapes for the cognitive and compassionate wisdom dimensions and an intermediate shape for the reflective dimension. Whereas Webster and colleagues (2014) found that the relation between age and wisdom, assessed by the SAWS, followed an inverse U-curve for the composite wisdom measure and four of its subscales (with the exception of reminiscence), the shapes for the three wisdom dimensions all varied in the present study. As predicted in Hypothesis 2 and the “fluid intelligence” view (Sternberg, 2005), average scores on the cognitive wisdom dimension were relatively stable from age 18 until midlife and then declined, which might have been due to decreases in fluid intelligence (Kievit et al., 2016), increases in rigidity or dogmatism (Meacham, 1990; Schultz & Searleman, 2002), and a decrease in openness to experiences, which Webster and colleagues (2014) also reported. Average scores on the reflective wisdom dimension first increased until midlife, when individuals might have a greater need to perceive phenomena and events from different perspectives to understand and manage their environment effectively, stayed relatively stable between 40 and 70, and increased again after age 70. This corroborates Hypothesis 3 and lends credence to the possibility that older adults might be impelled to engage in life review and (self-)reflection to overcome despair and master Erikson’s (1963) psychosocial task of achieving ego integrity when death looms ever closer on the horizon. Although research has found that life review is not limited to older age groups, the purpose of life review and reflection differs across the life course with death preparation being particularly relevant in old age (Staudinger, 2001). In direct opposition to the cognitive wisdom dimension, average scores on the compassionate wisdom dimension decreased slightly until midlife and then increased again as expected in Hypothesis 4. This pattern fits with Erikson’s (1963) psychosocial developmental pathway toward generativity at midlife, when self-centered ambitions and impulses are replaced by a concern for the welfare of the next generation and an increase in compassion for others. Greater average scores on the reflective and compassionate wisdom dimensions with advanced age are also compatible with an increased need for emotional regulation in old age and consistent with prior research (Chopik et al., 2017; Neff & Pommier, 2013; Scheibe & Carstensen, 2010). Educational attainment was positively related to wisdom and the cognitive and reflective wisdom dimensions and affected the shape of the age and wisdom relationship, including its three dimensions. For participants with an elementary education and men with an intermediate education, average wisdom scores first increased until midlife and then declined again, with a sharper decline for those with an elementary education. Interestingly, for women with an intermediate education, wisdom tended to increase until around age 60 and then decline only slightly. For participants with a high school or university education, age was unrelated to wisdom (but not positively related as predicted in Hypothesis 5), indicating that a higher education might have preserved an openness to experiences and prevented mental rigidity and dogmatism and, therefore, a loss of wisdom during the later years of life. A very similar pattern was observed for the association between age and the cognitive wisdom dimension, except that for participants with a university education, the relation also became curvilinear, with average scores increasing until midlife followed by a gradual decline but still higher average scores than any other educational group until around age 70. Women with an intermediate education tended to have scores on wisdom and the cognitive wisdom dimension similar to men with an intermediate education and those with an elementary education up to the age of about 35 and then tended to become more similar to those with a higher education after the age of about 45. It is likely that this reflects cohort differences as older women in Germany were probably encouraged to pursue an intermediate rather than a high school or university education even if they had the academic aptitude to obtain a higher educational degree (Breen, Luijkx, Müller, & Pollak, 2010). Hypothesis 6 that the expected growth in the reflective wisdom dimension in old age and the compassionate wisdom dimension after midlife would occur only for participants with a high school or university education was supported, but only for ages up to 70. Hence, participants with a higher education might be more likely to engage in life review and reflection in old age and generativity in midlife than those with a lower education. Beyond the age of 70, however, reflective dimension scores tended to increase for those with an intermediate education, and compassionate dimension scores tended to decline for those with a high school education. Yet, because the number of respondents over the age of 70 was relatively small, it is not clear how reliable these results are. The major limitations of the study are the cross-sectional and nonrepresentative nature of the data. Participants tended to be more educated than the general population of Germany and older adults over the age of 70 were underrepresented. Another limitation is that participants were aware that they took a “wisdom test,” which might have biased their answers in a presumably “wiser” direction, although Figures 1 and 2 show that mean values of wisdom scores and its three dimensions were not excessively inflated and comparable to other studies that used the 3D-WS without informing participants that the questionnaire contained a wisdom scale (e.g., Ardelt, 2010; Ardelt & Jeste, In press; Glück et al., 2013; Zacher, McKenna, & Rooney, 2013). Still, to be confident of the validity of the results, the study needs to be replicated with a large representative sample without informing participants that the study attempts to measure wisdom. Yet, the association between age and wisdom can ultimately only be tested with longitudinal data for different cohorts as cohort effects might be partly responsible for the results. The older cohorts in the sample experienced WWII and its aftermath in Germany and might have been more concerned with survival than the pursuit of wisdom (Koltko-Rivera, 2006). Despite its limitations, this study is the first with a sample large enough to analyze the relation between age and three-dimensional wisdom separately by educational groups. Although decreases in wisdom have been found during later life, the research indicated that the compassionate and reflective dimensions of wisdom might increase after midlife or during old age, respectively, for those with a higher education, which might be related to a growing need for emotion regulation (Carstensen & Mikels, 2005; Charles & Carstensen, 2009; Scheibe & Carstensen, 2010) and the successful mastery of Erikson’s (1963) psychosocial crises of generativity versus stagnation at midlife and ego integrity versus despair in old age. According to Erikson’s stage model, the resolution of these crises is facilitated by the successful mastery of four psychosocial crises in childhood and two in adolescence and emerging adulthood. Hence, positive human development in early life might lay the groundwork for greater wisdom later in life. Conflict of Interest None reported. Acknowledgements An earlier version of this study was presented at the 2016 European Sociological Association Research Network on Ageing in Europe Conference, Frankfurt, Germany. We like to thank the GEO technical team for providing the data and Lisa Christiansen for helping to clean up the data. References Aiken, L. S., & West, S. G. ( 1991). Multiple regression: Testing and interpreting interactions . Thousand Oaks, CA: Sage. Ardelt, M. ( 2000). Intellectual versus wisdom-related knowledge: The case for a different kind of learning in the later years of life. Educational Gerontology: An International Journal of Research and Practice , 26, 771– 789. doi:10.1080/036012700300001421 Google Scholar CrossRef Search ADS   Ardelt, M. ( 2003). Empirical assessment of a three-dimensional wisdom scale. Research on Aging , 25, 275– 324. doi:10.1177/0164027503025003004 Google Scholar CrossRef Search ADS   Ardelt, M. ( 2010). Are older adults wiser than college students? A comparison of two age cohorts. Journal of Adult Development , 17, 193– 207. doi:10.1007/s10804-009-9088-5 Google Scholar CrossRef Search ADS   Ardelt, M., & Jeste, D. ( In press). Wisdom and hard times: The ameliorating effect of wisdom on the negative association between adverse life events and well-being. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences . doi:10.1093/geronb/gbw137 Baltes, P. B., & Smith, J. ( 2008). The fascination of wisdom: Its nature, ontogeny, and function. Perspectives on Psychological Science: A Journal of the Association for Psychological Science , 3, 56– 64. doi:10.1111/j.1745-6916.2008.00062.x Google Scholar CrossRef Search ADS PubMed  Baltes, P. B., & Staudinger, U. M. ( 2000). Wisdom. A metaheuristic (pragmatic) to orchestrate mind and virtue toward excellence. The American Psychologist , 55, 122– 136. doi:10.1037/0003-066X.55.1.122 Google Scholar CrossRef Search ADS PubMed  Bergsma, A., & Ardelt, M. ( 2012). Self-reported wisdom and happiness: An empirical investigation. Journal of Happiness Studies , 13, 481– 499. doi:10.1007/s10902-011-9275-5 Google Scholar CrossRef Search ADS   Blanchard-Fields, F., & Norris, L. ( 1995). The development of wisdom. In M. A. Kimble, S. H. McFadden, J. W., J. J. Ellor, & Seeber (Eds.), Aging, spirituality, and religion. A handbook  (pp. 102– 118). Minneapolis, MN: Fortress Press. Breen, R., Luijkx, R., Müller, W., & Pollak, R. ( 2010). Long-term trends in educational inequality in Europe: Class inequalities and gender differences. European Sociological Review , 26, 31– 48. doi:10.1093/esr/jcp001 Google Scholar CrossRef Search ADS   Carstensen, L. L., & Mikels, J. A. ( 2005). At the intersection of emotion and cognition: Aging and the positivity effect. Current Directions in Psychological Science , 14, 117– 121. doi:10.1111/j.0963-7214.2005.00348.x Google Scholar CrossRef Search ADS   Charles, S. T., & Carstensen, L. L. ( 2009). Social and emotional aging. Annual Review of Psychology , 61, 383– 409. doi:10.1146/annurev.psych.093008.100448 Google Scholar CrossRef Search ADS   Cheraghi, F., Kadivar, P., Ardelt, M., Asgari, A., & Farzad, V. ( 2015). Gender as a moderator of the relation between age cohort and three-dimensional wisdom in Iranian culture. International Journal of Aging and Human Development , 81, 3– 26. doi:10.1177/0091415015616394 Google Scholar CrossRef Search ADS PubMed  Chopik, W. J., O’Brien, E., & Konrath, S. H. ( 2017). Differences in empathic concern and perspective taking across 63 countries. Journal of Cross-Cultural Psychology , 48, 23– 38. doi:10.1177/0022022116673910 Google Scholar CrossRef Search ADS   Clayton, V. P., & Birren, J. E. ( 1980). The development of wisdom across the life-span: A reexamination of an ancient topic. In Baltes P. B. & Brim O. G. (Eds.), Life-span development and behavior  (Vol. 3, pp. 103– 135). New York, NY: Academic Press. Erikson, E. H. ( 1959). Identity and the life cycle . New York, NY: International Universities Press, Inc. Erikson, E. H. ( 1963). Childhood and society . New York, NY: Norton. Ferrari, M., Kahn, A., Benayon, M., & Nero, J. ( 2011). Phronesis, sophia, and hochma: Developing wisdom in Islam and Judaism. Research in Human Development , 8, 128– 148. doi:10.1080/15427609.2011.568869 Google Scholar CrossRef Search ADS   Fischer, A. ( 2015). Wisdom - the answer to all the questions really worth asking. International Journal of Humanities and Social Science , 5, 73– 83. Glück, J. ( 2016). Weisheit: Die 5 Prinzipien des gelingenden Lebens (Wisdom: The 5 principles of a succeeding life) . München, Germany: Kösel. Glück, J., König, S., Naschenweng, K., Redzanowski, U., Dorner, L., Straßer, I., & Wiedermann, W. ( 2013). How to measure wisdom: Content, reliability, and validity of five measures. Frontiers in Psychology , 4, 405. doi:10.3389/fpsyg.2013.00405 Google Scholar CrossRef Search ADS PubMed  Horn, J. L., & Cattell, R. B. ( 1966). Age differences in primary mental ability factors. Journal of Gerontology , 21, 210– 220. doi:10.1093/geronj/21.2.210 Google Scholar CrossRef Search ADS PubMed  Kaufman, A. S. ( 2001). WAIS-III IQs, Horn’s theory, and generational changes from young adulthood to old age. Intelligence , 29, 131– 167. doi:10.1016/s0160-2896(00)00046-5 Google Scholar CrossRef Search ADS   Kekes, J. ( 1983). Wisdom. American Philosophical Quarterly , 20, 277– 286. Kievit, R. A., Davis, S. W., Griffiths, J., Correia, M. M., Cam-Can, & Henson, R. N. ( 2016). A watershed model of individual differences in fluid intelligence. Neuropsychologia , 91, 186– 198. doi:10.1016/j.neuropsychologia.2016.08.008 Google Scholar CrossRef Search ADS PubMed  Kitchener, K. S., & Brenner, H. G. ( 1990). Wisdom and reflective judgment: Knowing in the face of uncertainty. In Sternberg R. J. (Ed.), Wisdom: Its nature, origins, and development  (pp. 212– 229). New York, NY: Cambridge University Press. Google Scholar CrossRef Search ADS   Kohn, M. L., & Schooler, C. ( 1983). Work and personality: An inquiry into the impact of social stratification . Norwood, NJ: Ablex. Kohn, M. L., Zaborowski, W., Janicka, K., Mach, B. W., Khmelko, V., Slomczynski, K. M., … Podobnik, B. ( 2000). Complexity of activities and personality under conditions of radical social change: A comparative analysis of Poland and Ukraine. Social Psychology Quarterly , 63, 187– 208. doi:10.2307/2695868 Google Scholar CrossRef Search ADS   Koltko-Rivera, M. E. ( 2006). Rediscovering the later version of Maslow’s hierarchy of needs: Self-transcendence and opportunities for theory, research, and unification. Review of General Psychology , 10, 302– 317. doi:10.1037/1089-2680.10.4.302 Google Scholar CrossRef Search ADS   Kramer, D. A. ( 1990). Conceptualizing wisdom: The primacy of affect-cognition relations. In Sternberg R. J. (Ed.), Wisdom: Its nature, origins, and development  (pp. 279– 313). New York, NY: Cambridge University Press. Google Scholar CrossRef Search ADS   Levenson, M. R., Jennings, P. A., Aldwin, C. M., & Shiraishi, R. W. ( 2005). Self-transcendence: Conceptualization and measurement. International Journal of Aging & Human Development , 60, 127– 143. doi:10.2190/XRXM-FYRA-7U0X-GRC0 Google Scholar CrossRef Search ADS PubMed  Meacham, J. A. ( 1990). The loss of wisdom. In Sternberg R. J. (Ed.), Wisdom: Its nature, origins, and development  (pp. 181– 211). New York, NY: Cambridge University Press. Google Scholar CrossRef Search ADS   Murphy, S. L., Xu, J., & Kochanek, K. D. ( 2013). Deaths: final data for 2010. National Vital Statistics Reports: From the Centers for Disease Control and Prevention, National Center for Health Statistics, National Vital Statistics System , 61, 1– 117. Google Scholar PubMed  Neff, K. D., & Pommier, E. ( 2013). The relationship between self-compassion and other-focused concern among college undergraduates, community adults, and practicing meditators. Self and Identity , 12, 160– 176. doi:10.1080/15298868. 2011.649546 Google Scholar CrossRef Search ADS   Pasupathi, M., & Staudinger, U. M. ( 2001). Do advanced moral reasoners also show wisdom? Linking moral reasoning and wisdom-related knowledge and judgement. International Journal of Behavioral Development , 25, 401– 415. doi:10.1080/016502501316934833 Google Scholar CrossRef Search ADS   Pasupathi, M., Staudinger, U. M., & Baltes, P. B. ( 2001). Seeds of wisdom: Adolescents’ knowledge and judgment about difficult life problems. Developmental Psychology , 37, 351– 361. Google Scholar CrossRef Search ADS PubMed  Richardson, M. J., & Pasupathi, M. ( 2005). Young and growing wiser: Wisdom during adolescence and young adulthood. In R. J. Sternberg & J. Jordan (Eds.), A handbook of wisdom. Psychological perspectives  (pp. 139– 159). New York, NY: Cambridge University Press. Google Scholar CrossRef Search ADS   Rokeach, M. ( 1954). The nature and meaning of dogmatism. Psychological Review , 61, 194– 204. doi:10.1037/h0060752 Google Scholar CrossRef Search ADS PubMed  Scheibe, S., & Carstensen, L. L. ( 2010). Emotional aging: Recent findings and future trends. The Journals of Gerontology: Series B: Psychological Sciences and Social Sciences , 65, 135– 144. doi:10.1093/geronb/gbp132 Google Scholar CrossRef Search ADS   Schultz, P. W., & Searleman, A. ( 2002). Rigidity of thought and behavior: 100 years of research. Genetic, Social, and General Psychology Monographs , 128, 165– 207. Google Scholar PubMed  Staudinger, U. M. ( 2001). Life reflection: A social-cognitive analysis of life review. Review of General Psychology , 5, 148– 160. doi:10.1037/1089-2680.5.2.148 Google Scholar CrossRef Search ADS   Staudinger, U. M., & Glück, J. ( 2011). Psychological wisdom research: Commonalities and differences in a growing field. Annual Review of Psychology , 62, 215– 241. doi:10.1146/annurev.psych.121208.131659 Google Scholar CrossRef Search ADS PubMed  Staudinger, U. M., & Kunzmann, U. ( 2005). Positive adult personality development: Adjustment and/or growth? European Psychologist , 10, 320– 329. doi:10.1027/1016-9040.10.4.320 Google Scholar CrossRef Search ADS   Sternberg, R. J. ( 2005). Older but not wiser? The relationship between age and wisdom. Ageing International , 30, 5– 26. doi:10.1007/BF02681005 Google Scholar CrossRef Search ADS   Takahashi, M., & Overton, W. F. ( 2002). Wisdom: A culturally inclusive developmental perspective. International Journal of Behavioral Development , 26, 269– 277. doi:10.1080/01650250143000139 Google Scholar CrossRef Search ADS   Thomas, M. L., Bangen, K. J., Ardelt, M., & Jeste, D. V. ( 2017). Development of a 12-item abbreviated Three-Dimensional Wisdom Scale (3D-WS-12): Item selection and psychometric properties. Assessment , 24, 71– 82. doi:10.1177/1073191115595714 Google Scholar CrossRef Search ADS PubMed  Walsh, R. ( 2015). What is wisdom? Cross-cultural and cross-disciplinary syntheses. Review of General Psychology , 19, 278– 293. doi:10.1037/gpr0000045 Google Scholar CrossRef Search ADS   Webster, J. D. ( 2007). Measuring the character strength of wisdom. International Journal of Aging and Human Development , 65, 163– 183. doi:10.2190/AG.65.2.d Google Scholar CrossRef Search ADS PubMed  Webster, J. D., Westerhof, G. J., & Bohlmeijer, E. T. ( 2014). Wisdom and mental health across the lifespan. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 69, 209– 218. doi:10.1093/geronb/gbs121 Google Scholar CrossRef Search ADS PubMed  Wink, P., & Helson, R. ( 1997). Practical and transcendent wisdom: Their nature and some longitudinal findings. Journal of Adult Development , 4, 1– 15. doi:10.1007/BF02511845 Google Scholar CrossRef Search ADS   Zacher, H., McKenna, B., & Rooney, D. ( 2013). Effects of self-reported wisdom on happiness: Not much more than emotional intelligence? Journal of Happiness Studies , 14, 1697– 1716. doi:10.1007/s10902-012-9404-9 Google Scholar CrossRef Search ADS   © The Author(s) 2018. 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 Journals of Gerontology Series B: Psychological Sciences and Social SciencesOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off