Despite the impressive collection of research findings for personality stability using the five-factor model, as we can see from the previous discussion there is growing evidence for personality change. Ursula Staudinger and colleagues have a perspective that reconciles these differences (Staudinger & Kunzman, 2005; Staudinger & Kessler, 2008). They suggest that personality takes on two forms: adjustment and growth. Adjustment involves developmental changes in terms of their adaptive value and functionality such as funcitoning effectively within society and how personality contributes to everyday life running smoothly. Personality growth refers to ideal end states such as increased self-transcendence, wisdom, and integrity. Examples of this will be discussed later on including Eirkson’s theory and Loevinger’s theory. Both of these types of personality dimen­sions interact in that growth cannot occur without adjustment. However, Staudinger argues that while growth in terms of ideal end states does not neces­sarily occur in everyone, as it is less easily acquired, strategies for adjustment develop across the latter half of the life span. We can take this framework and interpret stability and change in the Big Five personality factors.

First, the most current consensus of change in the Big Five with increasing age is the absence of neuroticism and the presence of agreeableness and conscientiousness. These traits are associated with personality adjustment. This pattern of traits is related to becoming emotionally less volatile and more attuned to social demands and social roles (Staudinger & Kunzmann, 2005). These character­istics prepare adults to tackle developmental tasks and adult roles. In fact, they allow older adults to maintain and regain levels of well-being in the face of loss, threats, and challenges in life.

Studies also show a decrease in openness to new experiences with increasing age (e. g., Helson & Kwan, 2000; Roberts et al., 2006; Srivastava et al.,

2003) . Staudinger argues that openness to experi­ence is related to personal maturity in that it is highly correlated with ego development, wisdom, and emotional complexity. Evidence suggests that these three aspects of personality (ego level, wisdom, and emotional complexity) do not increase with age and may show decline (Dorner & Staudinger, 2005; Labouvie-Vief & Medler, 2002). Staudinger con­cludes that personal growth in adulthood appears to be rare rather than normative.

In sum, there appears to be increases in adjust­ment aspects of personality with increasing age and it could be normative. However, indicators of per­sonality growth tend to show stability or decline. It is not that personality growth cannot occur in older age; it may simply take very special circumstances and environmental push to occur. It is important to keep in mind that personality-related adjustment continues to grow as we face ever-changing devel­opmental challenges and tasks to accomplish such as establishing a career, marriage, and family.

Other studies support Staudinger’s model of personality and illustrate how our techniques for assessing personality have changed. For example, Srivastava, John, Gosling, and Potter (2003) con­ducted a large Internet study examining the Big Five traits of more than 130,000 people ranging in age from 21 to 60 years. This study is a testament to how changes in our technology allow more in-depth analyses of larger samples of individuals. Not only has technology changed our techniques for assess­ment (using the Internet), but it also has changed our techniques for recruitment. For example, to attract a broad and diverse sample, these researchers used two types of web pages. One was a guide to personality instructing i ndividuals on the funda­mental dimensions of personality, and the second was a page headed “Find your Star Wars Twin” that included feedback about the characters from Star Wars with whom the participant was most similar. Needless to say they had no trouble obtaining a large sample of participants. They found that none of the Big Five personality traits remained stable after age 30. Supporting the research discussed earlier, conscientiousness showed the most change in early adulthood, a time when adults are advancing in
the workforce and forming intimate relationships. Agreeableness showed the largest changes in middle adulthood, when adults are typically caring for their children. The researchers concluded, as well, that it is important to link personality and environmental changes to one another.

Finally, it is important to consider not only age changes in personality traits such as neuroticism but also cohort differences. Twenge (2000) conducted an analysis across many longitudinal studies and found that Americans have shown higher levels of anxiety and neuroticism during recent decades. In other words, she found a cohort difference that may influence future observations of changes in neuroti – cism across adulthood. The average American child in the 1980s reported more anxiety than children in the 1950s. Perhaps increases in divorce rate could account for these findings. Twenge concludes that broad social trends may have an important impact on personality development, particularly in child­hood, that may have interesting effects on what we observe in adulthood.

Finally, we need to briefly consider two other longitudinal studies of dispositional traits that move beyond the Big Five factors.

The Berkeley Studies. Researchers in Berkeley, Cali­fornia, conducted one of the largest longitudinal studies on personality development. In this investi­gation, the parents of participants being studied in research on intellectual development were followed for roughly 30 years between ages 40 and 70 (Maas, 1985; Maas & Kuypers, 1974; Mussen, 1985). From the enormous amount of data gathered over the years, researchers were able to categorize men and women into subgroups based on their lifestyles (for instance, whether mothers were employed) and personality type. Based on the longitudinal follow-up data, gender differences were identified in terms of the best predictors of life satisfaction in old age. The data suggest that lifestyle during young adulthood is the better predictor of life satisfaction in old age for women but that personality is the better predictor for men (Mussen, 1985).

Additional analyses of the Berkeley data provide other insights into personality development. For

example, the transition to parenthood in young adulthood and family and work-related transitions (e. g., retirement) in later adulthood showed much more variability than stability. Again, the sociocul­tural context was important in determining under what conditions we see change and when we do not. These results also do not necessarily support the notion that personality becomes rigid in old age (Field & Millsap, 1991; Labouvie-Vief et al., 2000).

Women’s Personality Development during Adulthood.

Several longitudinal studies of women’s personality development have adopted a process approach in which they examine the interplay between social con­text and personality development (Helson & Moane, 1987; Helson et al., 1995; Roberts et al., 2002; Stewart et al., 2001). Helson and her colleagues followed the lives of women who chose a typcally feminine social clock (get married, have children, etc.) and examined how they adapted to the roles of wife and mother. This adaptation process was typically accompanied by a withdrawal from social life, the suppression of impulse and spontaneity, a negative self-image, and decreased feelings of competence. Of the women who adhered to the social clock, 20% were divorced between the ages of 28 and 35. However, of these divorced women, those who also had careers by the age of 28 were less respectful of norms and more rebellious toward what they experienced. Note that these women were not lower on femininity or on well-being; they were simply more independent and self-assertive than those who followed the social clock (Helson & Moane, 1987). Follow-ups showed that these independent women remained so and showed greater confidence, initiative, and forceful­ness than women who did not. In fact, women were better adjusted at age 52 than at age 21. They became less impulsive, more considerate of others, more organized, more complex, and better able to adapt to various settings (Roberts et al., 2002; Stewart et al.,

2001) . Overall Helson and her colleagues show that women’s personality change was systematic in both early and middle adulthood, yet changes were evi­dent in the context of specific changes in social roles and transitions in social contexts (Labouvie-Vief et al., 2000; Van Manen & Whitbourne, 1997).

Middle-aged women are becoming more active and visible in our society.

Critiques of the Five-Factor Model

In a major review of the literature, Block (1995) raises several concerns with the Costa and McCrae approach. Most of Block’s criticisms are based on perceived methodological problems, such as the way the dimensions were identified statistically and the way the questionnaire assessment was developed and used. This critique is based on the view that the statistical and empirical grounds on which the five – factor model is built are shaky. For example, Block argues that using laypeople to specify personality descriptors, the approach used to create the terms used in the five-factor model, is fraught with risk, chiefly due to the lack of any compelling scientific data to support such labeling. Thus Block argues that the wide acceptance of the five-factor model is premature and that considerably more research needs to be done. Block also argues in favor of personality research that takes into consideration the sociocultural context in which personality development occurs and the variability that occurs across the life course (Labouvie-Vief & Diehl, 1999; Labouvie-Vief et al., 2000). As we can see from the preceding studies, current research has followed suit with this criticism.

Finally, McAdams (1996, 1999) has raised addi­tional limitations of the five-factor model. He points out that any model of dispositional traits says noth­ing about the core or essential aspects of human nature. In contrast, theorists we will consider later,

such as Erikson and Loevinger, do discuss such core aspects, which cannot be translated into the language of dispositional traits. Second, disposi­tional traits rarely provide enough information about people so that accurate predictions can be made about how they will behave in a particular situation. Third, the assessment of dispositional traits generally fails to provide compelling explana­tions of why people behave the way they do. Fourth, dispositional traits are seen as independent of the context in which the individual operates. In other words, similar to Block’s objections, the five-factor model approach ignores the sociocultural context of human development. Fifth, the assessment of dispo­sitional traits reduces a person to a set of scores on a series of linear continua anchored by terms that are assumed to be both meaningful and opposite. Sixth, the assessment of dispositional traits through ques­tionnaires assumes that the respondent is able to take an objective, evaluative stance regarding his or her personal characteristics. McAdams’s criticisms nevertheless reflect the view that assessing disposi­tional traits has its place in personality research; his point is simply that dispositional traits should not be viewed as reflecting one’s entire personality.

As we might expect, the critiques themselves are controversial. For example, Costa and McCrae (1995, 1998) and Goldberg and Saucier (1995) point to flaws in Block’s argument, such as his overlooking of research favorable to the five-factor model. Arguments about the place for assessing dispositional traits within the study of personality reflect the biases of the authors. The controversy surrounding Costa and McCrae’s basic claim that dispositional traits typically remain stable in adult­hood will continue, and it is likely to result in more, and even better, research.

In fact, one of the most recent and exciting approaches to this controversy is to take an intrain­dividual perspective (Mroczek & Spiro, 2003). Mroczek and colleagues challenge the conclusions drawn from the typical longitudinal studies on s tability and change in personality by examining personality across the adult life span at the level of the individual. We describe this challenge in more detail in the Current Controversies feature.

324 CHAPTER 9

Conclusions about Dispositional Traits

What can we conclude from the research on the development of personality traits across adulthood? From both conceptual and empirical perspectives it seems that the idea that personality traits stop chang­ing at age 30 does not have uniform support. On the one hand, we have a definition of traits that requires stability. Costa and McCrae, among others, argue strongly for this position; they report that there is little evidence (and perhaps possibility) of change. On the other hand, Staudinger’s perspective, the Berkeley group, and Helson and colleagues argue for both change and stability. They say that at least some traits change, opening the door to personality development in adulthood. One partial resolution can be found if we consider how the research was done. Clearly, the overwhelming evidence sup­ports the view that personality traits remain stable throughout adulthood when data are averaged across many different kinds of people. However, if we ask about specific aspects of personality in very specific kinds of people, we are more likely to find some evidence of both change and stability.

The recent critiques of the five-factor model have created a climate in which more careful scrutiny of research on dispositional traits is likely to occur. More sophisticated statistical techniques are now being applied, and more attention is being paid to individual differences in change as we saw in the Current Controversies feature. Helson and colleagues (2002) suggest that we need to consider personality as consistent yet adaptive to changing biological and environmental contexts. These approaches to per­sonality help address many of the criticisms.

In addition, the idea that trait stability may not be fixed at age 30 invites the question as to whether periods beyond age 30 may be associated with greater trait stability (Roberts & Del Vecchio,

2000) . In fact, Roberts and Del Vecchio find that trait stability peaks in middle age when identity certainty (achieving a strong sense of identity, abil­ity to choose environments that fit well with one’s identity, and the ability to assimilate more expe­rience into one’s identity) is more likely to be achieved (Stewart et al., 2001). Future research needs to examine the extent to which many factors

CURRENT

CONTROVERSIES