Emotional Well-Being Following a Later Life Career Change: The Roles of Agency and Resources

Abstract

More individuals are attempting career changes in later life, as an increasing number of people face precarious retirement prospects. Although many of these older job seekers eventually find new livelihoods, little is known about their emotional well-being subsequent to these changes. Using the 2014 American Institute for Economic Research Older Worker Survey, we evaluate the contributions of demographic characteristics, agency, and resources when estimating three measures of emotional well-being following a successful later life career change (n = 337). We found that having financial resources during the career transition was associated with all three positive emotional outcomes, while family support and intentionality were also associated with positive emotions. Conversely, prior job prestige and purposeful job training had no relationships with subsequent well-being. These results suggest that later life career changes, despite their challenges, often result in positive emotional outlooks—for those who have the resources to support them.

Keywords

career change emotional well-being agency resources second careers

Introduction

Labor force participation rates among older adults have continued to increase, even as those among other age groups have either leveled off or declined (Toossi, 2013). A significant portion of this growth is the result of older adults delaying retirement due to (a) a decline in union jobs, (b) increasing lifetime employment insecurity, (c) concerns over inadequate savings, and (d) increasing health-care costs (Kalleberg, 2009; Wise, 2010). At the same time, the “graying of the workforce” has also brought about positive changes with respect to career and retirement choices (Alley & Crimmins, 2007), including a greater ability to make later life career changes.

Despite these societal transformations, the vast majority of research on career changes has focused on younger workers (Wang, Olson, & Shultz, 2012). Even studies that have called attention to later life career changes primarily examined the challenges and stress of these transitions, particularly as it relates to involuntary job loss (Bailey & Hansson, 1995; Gallo, Bradley, Siegel, & Kasl, 2000). This emphasis on the negative aspects of later life career changes ignores that the majority of older workers who attempt a career change will eventually succeed (Kreisberg, 2015). In addition, prior work has not adequately studied (a) how those who successfully navigate career changes retrospectively evaluate this major life change and (b) which individual-level attributes may be associated with postchange emotional well-being. Using a national panel survey of older workers, the purpose of this study is to test whether two personal domains (agency and resources) are associated with three measures of emotional well-being following a successful career change. In doing so, we highlight possible consequences of growing income inequality among older adults and consider the implications of intentionality when changing careers at older ages.

Later Life Career Changes

While baby boomers (those born between 1946 and 1964) continue to transition into traditional retirement ages (i.e., 65+), the proportion of older adults staying in the workforce past this age continues to rise. More specifically, the labor force participation rate among those aged 65 and over increased from 12% in 1990 to 18% in 2013 and is projected to reach 23% by 2022 (Toossi, 2013). Although research has traditionally focused on the adverse reasons for these changes (e.g., inadequate savings or rising health-care costs), some of this increase is due to more individuals proactively seeking nontraditional career paths in later life. For example, there has been a marked uptick in workers taking “bridge jobs”—often defined as short-term employment occurring between an individual’s primary career and complete labor force withdrawal (Shultz, 2003). Importantly, prior work suggests not only that these bridge jobs provide a means to make career changes in later life but also that some bridge jobs should be construed as second careers (Beehr & Bennett, 2015; Hill, Snell, & Sterns, 2015). Indeed, bridge jobs may be one reason why an increasing number of individuals are pursuing new full-time “second careers,” following 20+ years in the workforce (Wang et al., 2012). While only about half the rate of all adults, people making career changes during their later years are a growing percentage of working adults between ages 45 and 65, ranging from 2% to 3% annually between 2004 and 2012 (Kreisberg, 2015). This translates into approximately one to two million career changes in the United States per year and represents a group about which scholars know very little (Kanfer, 2015; Wang et al., 2012).

Although attempting a career change in later life can introduce a number of challenges and complexities (e.g., difficulty in obtaining a new job, a possible decrease in income, new career not matching expectations), there are numerous reasons why more individuals are choosing this path. For example, there is growing concern among some older adults about becoming permanently trapped in “precarious employment” that provides low income, little prestige, or negligible job security (Raymo, Warren, Sweeney, Hauser, & Ho, 2011). In addition, some older adults look for new careers as a source of personal fulfillment in later life, and working in “desirable jobs” has been linked to psychological and physical well-being among older workers (Herzog, House, & Morgan, 1991). Lastly, some careers or working conditions are not amenable to the needs of older adults, which could result in a number of negative physical and psychological consequences (Morelock, McNamara, & James, 2017). In total, these changes signal that some older adults are transitioning into new careers not only because of negative macrolevel economic conditions but also because of intentional self-improvement decisions (Heckhausen & Shane, 2015). These attitude shifts are supported by one survey finding that approximately one sixth of all workers aged 50 to 64 expect to work for pay in a completely different field after retiring from their current job (Ambrose, 2015).

Emotional Well-Being Following a Later Life Career Change

Recent research has suggested that the most valued component of successful aging is a general perception of well-being, including happiness and the ability to take part in new experiences (Kelly & Lazarus, 2015). Because later life career changes are increasingly feasible events, there is a growing need to evaluate satisfaction with these changes and to what extent these changes impact emotional well-being. Unfortunately, the few studies that have examined associations between emotional well-being and later life career transitions have primarily focused on the impact of job search attempts. That line of research has generally found that undergoing these endeavors—not surprisingly—presents a number of psychological challenges and perceived obstacles (Bailey & Hansson, 1995). In particular, less favorable mental health—including depression—has been associated with (a) attempting career changes that were a result of involuntary job loss and (b) extended periods of unemployment that often accompany career change attempts (Brewington & Nassar-McMillan, 2000; Gallo, Bradley, Teng, & Kasl, 2006; Zenger, Brähler, Berth, & Stöbel-Richter, 2011). While important, this line of research essentially ignores (a) the psychological outcomes of those who successfully navigate career changes, (b) the implications of variation in personal attributes among successful later life career changers, and (c) the consequences of career change intentionality (as opposed to involuntary job loss) on well-being. Put another way, once-perceived difficulties and hindrances that were present during the career change attempt may become unsubstantiated among those who eventually find new careers. If so, these challenging life transitions could result in rewarding new experiences that coincide with improved psychological well-being.

There are a number of ways to define or operationalize positive emotional well-being with respect to a successful career change. For example, individuals could reflect on whether or not they are happy with the change, or how their job satisfaction compares before and after the change. It could also include measures of psychological utility. For example, some adults proactively make career transitions searching for a personal “renewal”; hoping that a change will provide greater meaning, career engagement, and a sense of purpose (Wang et al., 2012). It is only after a completed career transition that individuals can reflect and assess whether this renewal actually occurred. No matter how emotional well-being is measured, there are likely a number of personal and psychological domains that are associated with increased (or decreased) chances of reporting more positive emotional outcomes. In this article, we test two of these—agency and resources.

Agency

One ongoing debate that lies at the intersection of social gerontology and organizational psychology concerns how much agency older adults have in planning their later life career trajectories and retirement (Ekerdt, 2009). In an appeal for more empirical research on this topic, Henkens (2015) provides an Agency within Structure framework for these late-career transitions. In this model, external structural pressures on individual-level agency come from three sources: institutional context (e.g., social security and pension systems), organizational forces (e.g., economic circumstances of the job sector), and households (e.g., family, health). Agency, under this framework, refers to (a) how individuals respond to these pressures and (b) the characteristics of those who successfully navigate these pressures (e.g., intentionality, personality, knowledge, skills; Henkens, 2015).

In addition to shaping the outcomes of later life career change attempts, agency also likely influences how individuals view or experience these changes. Take, for example, prior research that has found associations between involuntary retirement and decreases in life satisfaction (Dingemans & Henkens, 2014). If leaving a job is, instead, initiated from a place of greater agency (voluntarily), then the transition is likely to be viewed more positively—particularly once the transition is completed.

It is important to clarify that agency does not only relate to the instigation of career transitions (e.g., intentionally leaving a job). It also encompasses a number of proactive behaviors and choices during the subsequent job search that help individuals achieve their goals (Solberg, Good, Fischer, Brown, & Nord, 1995). For example, job candidates may return to school and take classes or obtain other additional certifications (increasing knowledge, per Henkens’ model) believing that it will help them become more competitive in the job market. In addition, positive job market results—even those presumably influenced by structure—may reinforce or add to an individual’s sense of agency. For example, securing a job in a new career with ease (e.g., first application, short search duration) may increase an individual’s self-esteem. Greater self-esteem (personality, per Henkens’ model), in turn, has been linked to greater career happiness (Baumeister, Campbell, Krueger, & Vohs, 2003).

Resources

It is well established that resource differences among older adults are associated with numerous measures of physical and psychological well-being (Ferraro & Shippee, 2009). Resources generally refer to characteristics of individuals that have personal/intrinsic value (e.g., health, self-esteem) or assets that help individuals achieve desired results (e.g., money, social support; Hobfoll, 2002). Within Henkens’ agency and structure paradigm detailed earlier, resources are unique in that they can represent both a component of structure and ways to mitigate structure. For example, family support resources are a component of household structural forces, while personal financial resources help allay the pressure of institutional forces (e.g., pension instability). Because resources can be conceptualized in numerous ways, it is beneficial in this context to operationalize them as distinct personal attributes that can shape individuals’ retirement and late-career trajectories (Koopmann & Wang, 2015).

To help conceptualize the complex relationships between resources, time, and late-career experiences, Wang et al. (2012) proposed a Resource-based Dynamic Model. This framework demonstrates how various aspects of career-oriented outcomes (e.g., satisfaction, stress) fluctuate over time, as a result of increases or decreases in resources. For example, career satisfaction would likely decrease following a job loss (e.g., a drop in financial resources), but other resources (e.g., family support) could compensate for this deficiency. This model also aims to signal late-career turning points by identifying instances where changes in personal resources subsequently lead to positive or negative career-related outcomes.

Similar to agency, the vast majority of past research linking resources to older-adult careers has focused on retirement or involuntary job loss (Riumallo-Herl, Basu, Stuckler, Courtin, & Avendano, 2014; Wang & Shultz, 2010). This is an important limitation because there are reasons to believe that differences in resources are also associated with career changes. For example, those with sufficient financial resources may be able to (a) more easily leave a career they are unsatisfied with or (b) have the luxury of preferring “fit” to salary when making a career change.

Along these same lines, social and family support has been linked to a number of positive emotional and psychological outcomes among older adults (Charles & Carstensen, 2010; Zhang & Li, 2011). This is not surprising because socioemotional selectivity theory (Carstensen, 1991) suggests that, as individuals get older, they tend to devote more emotional energy into fewer social relationships (i.e., family and long-term friends). Because of this, receiving social support from relatives may be critical in finding emotionally satisfying new careers. Notably, mechanisms linking both agency and resources to positive emotional outcomes likely differ by the type of emotion measured. For example, financial and social resources may give someone better opportunities to transition into a less stressful career, but these same assets may not necessarily influence whether or not an individual is ultimately happy with the change.

Present Study

The present study has two goals pertaining to individuals who have successfully navigated a later life career change. One, we test to what extent two domains (agency and resources) help explain measures of emotional well-being that refer to these changes. Two, we examine how these associations differ by the type of agency or resource an individual has. Drawing upon Henkens’ (2015) Agency within Structure theory during later career transitions, we expect that greater levels of agency will be associated with positive emotional well-being. We also employ Wang et al.’s (2012) Resource-based Dynamic Model of mid- and late-career change to posit that resources during a successful career transition are positively associated with more positive emotional outcomes subsequent to that transition.

Method

Data

To test our hypotheses, we used the 2014 Older Worker Survey (OWS), administered by the American Institute for Economic Research (AIER). This survey was conducted using GfK Inc.’s KnowledgePanel—an online, probability-based panel that consists of approximately 55,000 individuals aged 18 and over. The KnowledgePanel was created using random address-based sampling from U.S. postal service delivery files, drawing from a sample frame that covered approximately 97% of households (Dennis, 2010). Further, supplemental statistical tests and procedures were conducted to ensure population representativeness and consider self-selection bias (see Dennis, 2010 for detailed procedures). Using the total sample frame, a screening questionnaire was sent to 2,745 randomly selected individuals; 2,009 of which (73.2%) completed the screener. Of those respondents who completed the screener, 405 qualified for the OWS study (i.e., individuals were at least 47 years old and attempted a career change anytime after age 45). Because our research questions investigate emotions related to completed career changes, our final analytical sample (n = 337; 83.2%) necessarily includes only those who were successful in their attempt. Those who were not successful (n = 68) were given an alternative questionnaire that focused on possible reasons for the failure. To ensure that our results were generalizable to the U.S. population, all statistical models were estimated using inverse probability weights.

Using the OWS study to address our research questions provided two key benefits. One, it is a recently collected data set whose primary focus was to ascertain what skills, resources, and supports are used during later life career changes (Kreisberg, 2015). Two, the OWS study addresses some limitations of other surveys used for similar topics—namely, the Current Population Survey (CPS). For example, the CPS counts career changes only during the prior year, while OWS asks questions about career changes that happened at any time after age 45 (modal response was that career changes occurred 2 to 3 years prior to the survey). In addition, the OWS surveyed a significant number of older adults, while the CPS does not include any workers greater than 65. More specifically, OWS respondents’ ages ranged from 47 to 87 (M = 62.9), and a majority of the sample (63%) were aged 60 and older. Descriptive statistics for our analytical sample can be found in Table 1.

Table 1.

Descriptive Statistics; AIER Older Workers Survey: Successful Career Changers (n = 337).

	n	Percentage/M (SD)
Very happy with career change	182	55.7%
Stress decrease since career change	212	63.7%
Feels like “new person” since career change	240	71.9%
Age		62.9 (8.3)
Sex: Female	167	49.6%
Education
No college	96	28.5%
Some college	111	32.9%
College graduate	130	38.6%
Race
White	279	82.8%
Black	22	6.5%
Hispanic	17	5.0%
Other/Mixed race	19	5.7%
Prior job: high status	84	24.9%
Agency
Wanted to make change	201	59.6%
Not laid off from prior career	239	70.9%
Short job search	235	69.7%
Took additional classes	144	42.7%
Resources
Extra financial resources	142	42.1%
Family support	297	88.1%

Note. AIER = American Institute for Economic Research.

Measures: Emotional Well-Being Following a Career Change

We include all three measures of emotional well-being that were included in the OWS, all of which specifically referenced a completed career change. The first of these directly asked respondents “How happy would you say you are with the (career) change?” while the other two items asked whether respondents agreed or disagreed with the statements “My stress level decreased significantly since changing careers,” and “Emotionally, I feel like a new person since switching careers.”

Answers to all three survey questions included two positive responses and two negative responses. To focus on positive emotional outcomes and more easily interpret results, we dichotomized these measures. For the latter two, respondents who “completely” or “mostly” agreed with the statements on having less stress (64%) and feeling like a new person (72%) were coded as a 1. For the happiness item, we treated only those who were “very happy” with their career change as having positive emotional well-being (56%) because being “somewhat” happy with the change (35%) was an ambiguous affirmation, and virtually all respondents (91%) fell into one of these two categories.

Measures: Predictor Variables

We measured agency using all four related survey questions. The first of these—Wanted to make change (60%)—was determined by coding 1 for individuals who disagreed with the statement “I did not really want to make a career change, but because of financial or family reasons, I felt I had to.” The second item—Not laid off from prior career (75%)—was established by the question “Why did you leave your prior occupation?” For this item, all respondents were coded as 1 (75%), unless they reported that they were “laid off,” “transferred,” or left because the “business was closed.” The third item, Short job search (70%), was created by coding 1 for those who disagreed with the statement “I had to look for a long time before landing my new career.” The fourth and final agency item, Took additional classes (43%) was coded 1 if respondents reported taking any classes or undergoing formal training to prepare them for the career change. We consider the implication of treating these items as separate constructs—as opposed to using an index—in the Sensitivity Analyses section.

Resources were operationalized using two items. The first of these— Extra financial resources—was coded as 1 if a respondent agreed with the statement “I had extra financial resources which helped me to be able to make my career change” (42%). The second item—Family support was coded 1 if a respondent agreed with the statement “My family was very supportive of my career change” (88%). A correlation matrix of the agency and resource predictor variables can be found in Table 2. This analysis reveals that the agency items were either not correlated with each other or only slightly correlated with each other (r ranged between −.01 and .19), while the two resource items were, essentially, not correlated (r = .09).

Table 2.

Correlation Matrix of Agency and Resource Measures; AIER Older Workers Survey: Successful Career Changers (n = 337).

		1	2	3	4	5	6
1	Wanted to make change	1.00
2	Not laid off from prior career	0.19*	1.00
3	Short job search	0.18*	0.16*	1.00
4	Took additional classes	0.01	0.06	−0.01	1.00
5	Extra financial resources	0.10	−0.01	0.03	0.06	1.00
6	Family support	0.09	0.07	0.10	−0.02	0.09	1.00

Note. AIER = American Institute for Economic Research.

p < .05.

In our models, we also control for a number of individual-level demographic attributes that likely have associations with emotional well-being, career change, agency, and resources—sex, race, education, and former occupational status. For prior job status, we identified 10 jobs that are widely considered to be relatively high-prestige jobs (e.g., engineer, computer programmer, teacher) and coded these as 1 (25% of all respondents).

Analytic Strategy

We evaluate three sets of logistic regression models (A, B, C) using robust standard errors, with each set estimating one measure of career-oriented emotional well-being. Each of these sets consists of a three-model sequence (e.g., Models A1, A2, A3) that helps establish to what extent demographic characteristics, agency, and resources contribute to more positive emotional outcomes. The first of these (M1) is a baseline model that contains all posited demographic covariates (sex, education, race, and prior job status). M2 includes the demographic variables from M1 and adds the four agency measures. In doing so, it also tests whether any associations between demographic characteristics and well-being that were identified in M1 are explained by agency differences between individuals. M3 extends M2 by testing whether relationships between agency and emotional well-being are explained by having resources that may have supported high levels of agency. We consider the implication of treating our three dependent variables as separate emotional constructs (as opposed to using an index of career-oriented well-being) in the Sensitivity Analyses section.

There were very few missing data items: For each of the three dependent variables, answers were missing in 10 or fewer surveys. In addition, for each of the six independent variables used to operationalize agency and resources, there were five or fewer missing responses. Missing items were coded as 0 for these analyses, and a sensitivity check concluded that this assumption had no material effect on the estimates.

Results

Tables 3, 4, and 5 present results from the logistic regression models. To ease interpretation, results are displayed as exponentiated coefficients, representing proportional odds ratios. These numbers represent the lower (for ratios <1.0) or greater (for ratios >1.0) multiplicative odds of reporting a more positive emotional outcome associated with having that characteristic when compared with not having that characteristic.

Table 3.

Odds Ratios of Reporting “Very Happy” With the Career Change (Set A).

	MA-1		MA-2		MA-3
	OR	(SE)	OR	(SE)	OR	(SE)
Female (ref: Male)	2.36**	(0.64)	2.01**	(0.59)	1.98*	(0.59)
Education
No college^R	1.00	(0.00)	1.00	(0.00)	1.00	(0.00)
Some college	1.11	(0.38)	0.95	(0.37)	0.92	(0.36)
College graduate	1.16	(0.38)	0.94	(0.34)	0.95	(0.34)
Race
White^R	1.00	(0.00)	1.00	(0.00)	1.00	(0.00)
Black	2.37	(1.40)	1.82	(1.11)	1.43	(0.88)
Hispanic	2.22	(1.33)	2.23	(1.36)	2.44	(1.51)
Other/Mixed race	6.65***	(5.75)	7.15***	(6.89)	5.57#	(5.51)
Prior job: high status	1.08	(0.33)	1.21	(0.38)	1.23	(0.39)
Agency
Wanted to make change			3.14***	(0.92)	2.90***	(0.85)
Not laid off from prior career			1.71#	(0.55)	1.85*	(0.60)
Short job search			1.66#	(0.51)	1.73#	(0.54)
Took additional classes			1.47	(0.45)	1.51	(0.46)
Resources
Extra financial resources					1.97*	(0.59)
Family support					1.05	(0.47)
AIC	421		396		394
N	327		327		327

Note. R = reference category; AIC = Akaike Information Criterion.

#p ≤ .10. *p ≤ .05. **p ≤ .01. ***p ≤ .001.

Table 4.

Odds Ratios of Reporting Stress Decrease Since Changing Careers (Set B).

	MB-1		MB-2		MB-3
	OR	(SE)	OR	(SE)	OR	(SE)
Female (ref: Male)	1.28	(0.35)	1.24	(0.35)	1.20	(0.35)
Education
No college^R	1.00	(0.00)	1.00	(0.00)	1.00	(0.00)
Some college	2.03*	(0.71)	2.14*	(0.77)	2.07*	(0.76)
College graduate	0.97	(0.36)	0.97	(0.37)	0.93	(0.34)
Race
White^R	1.00	(0.00)	1.00	(0.00)	1.00	(0.00)
Black	1.63	(0.97)	1.76	(1.06)	1.28	(0.84)
Hispanic	1.91	(1.61)	2.07	(1.67)	2.84	(2.14)
Other/Mixed race	1.22	(0.86)	1.39	(0.95)	0.99	(0.73)
Prior job: high status	1.65	(0.59)	1.70	(0.60)	1.79#	(0.62)
Agency
Wanted to make change			1.27	(0.37)	1.07	(0.32)
Not laid off from prior career			1.13	(0.35)	1.20	(0.38)
Short job search			1.04	(0.33)	1.00	(0.31)
Took additional classes			0.76	(0.22)	0.78	(0.23)
Resources
Extra financial resources					2.65**	(0.82)
Family support					2.53*	(1.24)
AIC	424		430		414
N	333		333		333

Note. R = reference category; AIC = Akaike Information Criterion.

#p ≤ .10; *p ≤ .05; **p ≤ .01.

Table 5.

Odds Ratios of Reporting Feeling Like a “New Person” Since Changing Careers (Set C).

	MC-1		MC-2		MC-3
	OR	(SE)	OR	(SE)	OR	(SE)
Female (ref: Male)	1.64	(0.50)	1.27	(0.41)	1.24	(0.41)
Education
No college^R	1.00	(0.00)	1.00	(0.00)	1.00	(0.00)
Some college	2.64**	(0.99)	2.41*	(0.97)	2.33*	(0.96)
College graduate	1.84	(0.64)	1.58	(0.59)	1.54	(0.58)
Race
White^R	1.00	(0.00)	1.00	(0.00)	1.00	(0.00)
Black	4.59*	(3.37)	5.36*	(3.56)	4.85*	(3.54)
Hispanic	1.65	(1.14)	1.71	(1.10)	2.42	(1.84)
Other/Mixed race	0.80	(0.69)	0.75	(0.62)	0.42	(0.38)
Prior job: high status	1.47	(0.56)	1.84	(0.69)	1.91	(0.82)
Agency
Wanted to make change			1.67#	(0.50)	1.41	(0.46)
Not laid off from prior career			2.55**	(0.86)	3.10**	(1.13)
Short job search			1.71#	(0.55)	1.77#	(0.58)
Took additional classes			1.42	(0.48)	1.66	(0.63)
Resources
Extra financial resources					3.96***	(1.46)
Family support					3.53**	(1.52)
AIC	381		366		340
n	334		334		334

Note. R = reference category; AIC = Akaike Information Criterion.

#p ≤ .10. *p ≤ .05. **p ≤ .01. ***p ≤ .001.

Table 3 displays results for Model Set A (respondents reporting they were “very happy” with their career change). In MA-1, being female and reporting other/mixed race were both associated with this positive outcome, while in MA-2, only one agency variable—having wanted to make the change—was associated with happiness (OR = 3.14). In MA-3, we found that having financial resources during the transition was associated with being very happy with the change (OR = 1.97), but family support was not. Including these resource measures in MA-3 also revealed a new relationship between not being laid off (agency) and happiness (OR = 1.85).

Table 4 displays results for Model Set B (reporting a stress decrease after the career change). In MB-1, having some college education (using no college as a reference category) was associated with this decrease, while MB-2 indicated that none of the four agency items were associated with stress changes. In MB-3, however, both resource measures were associated with reduced stress (OR = 2.65 for financial resources; OR = 2.53 for family support).

Table 5 displays results for Model Set C (feeling like a “new person” after the career change). Similar to MB-1, having some college education was associated with increased odds of positive emotional well-being in MC-1. In addition, Black career changers were estimated to have much greater odds of feeling like a new person when compared with Whites. In MC-2, only one agency measure was related to feeling like a new person. In particular, those respondents who voluntarily left their prior career were estimated to have about 2.5 times greater odds of feeling like a new person when compared with those respondents who experienced involuntary job loss. Lastly, and similar to relationships found when estimating decreased stress (MB-3), both of the resource measures were associated with improved well-being in MC-3 (OR = 3.96 for financial resources and OR = 3.53 for family support).

Sensitivity Analyses

We estimated an additional set of models that employed an index of all three emotional well-being measures as the dependent variable (α = .63). Results from these models led us to similar conclusions as those presented earlier. Specifically, in M3, financial resources and family support were both related to this emotional well-being index, as were two of the four agency measures (wanting to make the change and not having been laid off). Because our main analyses found that these relationships differed by the measure of emotional well-being, we opted to present our results as three separate regressions, so that our conclusions are more conservative.

For a second supplementary analysis, we combined the four agency measures into one index. Employing this index would have led to results indicating that each additional point in the index was associated with increased odds of being very happy (OR = 1.99) and feeling like a new person (OR = 1.88). Because (a) this scale had low reliability (α = .31) and (b) the main analyses revealed these relationships are primarily attributable to two particular items in the index, we separated the four items when presenting our results.

Discussion

This article represents an initial step in answering calls for research to investigate the motivations, characteristics, and training of those making mid- and late-career transitions (Yoo & Lee, 2017). While some research has identified individual-level characteristics associated with instigating later life career changes (Kanfer, Chen, & Pritchard, 2012), this study evaluates how personal domains contribute to well-being subsequent to these changes. More specifically, we drew upon Henkens’ (2015) Agency within Structure framework and Wang et al.’s (2012) Resource-based Dynamic Model, expecting that both agency and resources are potentially important correlates of postchange well-being. Matching expectations, we found resources during a career change to be a salient predictor of positive emotional outcomes that reference that change. Conversely, and contrary to expectations, we found only limited evidence that agency during the job search was related to career-oriented emotional outcomes after completing that search.

Decades have passed since the U.S. business environment shifted from one of “career contracts” centered around organizations, to one of more unstable “protean careers” that place job market demands onto individuals (Hall, 2004). Because OWS respondents, on average, entered the workforce around 1970, they represent a group that experienced this economic transition firsthand. This societal shift has resulted in workers necessarily taking ownership of their careers, including strategies (e.g., acquiring new job skills, ensuring career self-direction, maintaining an understanding of the job market) captured by aspects of agency we measure in this article (Hall & Mirvis, 1996). Our analysis found that, generally, two (of four) agency items were related to more positive career-related emotions. One characteristic these two items share is that both revolve around intentionality when attempting the career change. For example, our estimates indicate that wanting to make a career change (as opposed to responding to external pressure) was associated with being very happy with that change. Because this survey question was specifically framed with respect to household structure forces, our results provide some evidence that agency—as presented in Henkens’ (2015) model—may help in understanding emotional well-being related to these transitions.

We also found that a second type of intentionality—not being laid off—was associated with two of the three positive outcomes. Together, these results suggest that the internal impetus behind later life career moves may have lasting consequences. For example, prior research has found that older adults are more likely than younger adults to harness intrinsic motivation, as it pertains to career goals and rewards (Kooij, De Lange, Jansen, Kanfer, & Dikkers, 2011). Acting on these motives, in turn, may lead to greater career-oriented satisfaction in future years. These findings are particularly notable because later life career changers with intentionality are an understudied group (Barclay, Stoltz, & Chung, 2011).

In contrast to the two agency items that revolve around intentionality, we found no relationships between the other two agency items (acquiring skills/knowledge; short job search) and well-being. Although more research is needed, these results suggest that these types of agency under Henkens’ framework (proficiency) may have limited importance with respect to career-oriented emotional well-being. These null findings also point to one way the role of agency may differ between younger and older career changers. That is, prior studies have found that knowledge and skills were both associated with career-oriented emotions among younger adults (Wise & Millward, 2005).

Unlike relationships between agency and emotional well-being, associations between resources and emotional well-being were much more consistent across outcomes. For example, those reporting extra financial resources during a career transition were estimated to have between two and four times greater odds (depending on the outcome) of reporting more positive career-oriented well-being than those who made the transition without these resources. Because the ability to even attempt a career change is likely linked to income and savings (Greller & Simpson, 1999), this presents a possible “double jeopardy” for financially disadvantaged older individuals considering such a change. One reason this may be the case is that extra financial support allows job seekers to be more selective, helping individuals land a job that best matches their needs. Interestingly, about 60% of OWS respondents reported not having extra financial resources—which is approximately the same percentage of U.S. working households age 55 to 64 that report inadequate retirement savings (Rhee, 2013).

The second resource measure—family support—was also associated with improved well-being (for two of the three measures). Although we are not familiar with any research linking family relationships to later life career changes, our findings are not overly surprising. That is, family support is generally associated with positive emotional well-being among older adults (Chen & Feeley, 2014), and social support has been found to protect older individuals from various forms of psychological distress (Morin & Midlarsky, 2016). Interestingly, the vast majority of OWS career changers reported family support, with 88% of respondents attesting to their assistance. This high percentage only serves to emphasize the psychological and occupational implications of those who do not have this cooperation. One possibility for future research is to examine family structures and family relationships among older adults who do not feel supported by their family during difficult transitions. For example, it may be that those who did not report family support are more likely to have lost a spouse or live far from adult children, both of which presumably have implications for work, retirement, and well-being.

While not a primary purpose of this study, we also examined relationships between basic demographic characteristics and career-oriented well-being after a late-life career change. A notable finding from these analyses was that—compared with not having attended college—completing some college (but not graduating college) was associated with better outcomes for two of the three measures. A more detailed data set is required to understand why this may be the case, but one possibility is that those with only some college education felt “under placed” at their prior job. With additional work experience and wisdom, this group may benefit most from later life career moves.

Limitations and Conclusion

Although the OWS is the first study we are aware of that was specifically created to examine later life career changes, its use is accompanied by several important limitations. One, because the OWS is a cross-sectional survey, it relies on respondent recall to answer questions about prior work status, prior events, prior agency, and prior support. These responses were based upon events that happened—on average—between 2 and 3 years prior to the survey. That said, this limitation is likely ameliorated by employing straightforward questions that are, presumably, easy to recall (e.g., “Why did you leave your prior occupation?”; Shultz, Taylor, & Morrison, 2003). Two, we were limited to the three questions of career-oriented emotional well-being that were included in the survey. There are numerous other ways to measure these outcomes (e.g., I feel close to people at my new job, I deal with problems better at my new job), and other constructs may have different relationships with both agency and resources than those identified in this article. .

Three, the data set and our conclusions would have benefitted by asking respondents for additional detail on a few particular survey questions. For example, it was not clear exactly what “financial resources” entails. While it likely represents extra savings, it may also represent home equity or money gifted by family members. Additional information would also have been beneficial with respect to respondents’ family structure—including current marital status and sources of emotional support. For example, future studies could ask which family members (e.g., spouse, children, and siblings) provided the most support during a career change and consider whether relationships between family support and emotional well-being depend on who supplied it.

In conclusion, as the average retirement age continues to rise, older adults are increasingly attempting career changes and succeeding (Loi & Shultz, 2007). Viewed skeptically, this is one symptom of a challenging employment environment and changing workforce dynamics at the national and global levels. More optimistically, these changes are consequences of increasing longevity and greater freedom for individuals to shape their work trajectories. Research suggests that decisions made during one’s primary career (e.g., concurrently working part-time jobs, proactively seeking developmental experiences) may predict both bridge jobs and participation in second careers (Hill et al., 2015). With increasing acceptance of later life career changes and a growing recognition of their possible benefits, there may be advantages to workers planning for these changes decades in advance.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Eric M. Vogelsang, PhD, is an assistant professor of sociology at California State University, San Bernardino, where he serves as the Director of the Center on Aging. Dr. Vogelsang primarily teaches classes in social gerontology and medical sociology. His prior work has been published in the Journal of Gerontology: Social Sciences and The Gerontologist.

Kenneth S. Shultz, PhD, is a professor of organizational psychology at California State University, San Bernardino. He has published several books (e.g., Aging and Work in the 21st Century and Mid and Late Career Issues: An Integrative Perspective) and numerous articles. Dr. Shultz primarily teaches classes in organizational psychology, research methods, and statistics.

Deborah A. Olson, PhD, is an associate professor of management and leadership at the University of La Verne (ULV). Her prior work has been published in a variety of journals. She is also a coauthor of Mid and Later Career Issues: An Integrative Perspective. At ULV, Dr. Olson primarily teaches classes in leadership, human resource management, and organizational behavior.

References

Alley

Crimmins

(2007) The demography of aging and work. In: Shultz

K. S.

Adams

G. A.

(eds) Aging and work in the 21st century, New York, NY: Psychology Press, pp. 7–23.

Ambrose, E. (2015, September 8). AARP survey: Today’s ‘retirement’ dream often includes a job. Retrieved from http://blog.aarp.org/2015/09/08/aarp-survey-todays-retirement-dream-often-includes-a-job/.

Bailey

L. L.

Hansson

R. O.

(1995) Psychological obstacles to job or career change in late life. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 50(6): P280–P288.

Barclay

S. R.

Stoltz

K. B.

Chung

Y. B.

(2011) Voluntary midlife career change: Integrating the transtheoretical model and the life-span, life-space approach. The Career Development Quarterly 59(5): 386–399.

Baumeister

R. F.

Campbell

J. D.

Krueger

J. I.

Vohs

K. D.

(2003) Does high self-esteem cause better performance, interpersonal success, happiness, or healthier lifestyles? Psychological Science in the Public Interest 4(1): 1–44.

Beehr

T. A.

Bennett

M. M.

(2015) Working after retirement: Features of bridge employment and research directions. Work, Aging and Retirement 1(1): 112–128.

Brewington

J. O.

Nassar-McMillan

(2000) Older adults: Work-related issues and implications for counseling. The Career Development Quarterly 49(1): 2–15.

Carstensen

L. L.

(1991) Selectivity theory: Social activity in life-span context. Annual Review of Gerontology and Geriatrics 11: 195–217.

Charles

Carstensen

L. L.

(2010) Social and emotional aging. Annual Review of Psychology 61: 383–409.

10.

Chen

Feeley

T. H.

(2014) Social support, social strain, loneliness, and well-being among older adults an analysis of the health and retirement study. Journal of Social and Personal Relationships 31(2): 141–161.

11.

Dennis

J. M.

(2010) Knowledgepanel®: Processes & procedures contributing to sample representativeness & tests for self-selection bias, Menlo Park, CA: Knowledge Networks.

12.

Dingemans

Henkens

(2014) Involuntary retirement, bridge employment, and satisfaction with life: A longitudinal investigation. Journal of Organizational Behavior 35(4): 575–591.

13.

Ekerdt

D. J.

(2009) Frontiers of research on work and retirement. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 65B(1): 69–80.

14.

Ferraro

K. F.

Shippee

T. P.

(2009) Aging and cumulative inequality: How does inequality get under the skin? The Gerontologist 49(3): 333–343.

15.

Gallo

W. T.

Bradley

E. H.

Siegel

Kasl

S. V.

(2000) Health effects of involuntary job loss among older workers findings from the health and retirement survey. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 55(3): S131–S140.

16.

Gallo

W. T.

Bradley

E. H.

Teng

H.-M.

Kasl

S. V.

(2006) The effect of recurrent involuntary job loss on the depressive symptoms of older us workers. International Archives of Occupational and Environmental Health 80(2): 109–116.

17.

Greller

M. M.

Simpson

(1999) In search of late career: A review of contemporary social science research applicable to the understanding of late career. Human Resource Management Review 9(3): 309–347.

18.

Hall

D. T.

(2004) The protean career: A quarter-century journey. Journal of Vocational Behavior 65(1): 1–13.

19.

Hall

D. T.

Mirvis

P. H.

(1996) The new protean career: Psychological success and the path with a heart. In: Hall

D. T.

(ed.) The career is dead: Long live the career, San Francisco, CA: Jossey-Bass, pp. 15–45.

20.

Heckhausen

Shane

(2015) How individuals navigate social mobility: Changing capacities and opportunities in careers across adulthood. In: Finkelstein

Truxillo

Fraccaroli

Kanfer

(eds) Facing the challenges of a multi-age workforce: A use-inspired approach, New York, NY: Routledge, pp. 313–320.

21.

Henkens

(2015) Labor force transitions in late life: Between agency and structure. In: Finkelstein

Truxillo

Fraccaroli

Kanfer

(eds) Facing the challenges of a multi-age workforce: A use-inspired approach, New York, NY: Routledge, pp. 321–329.

22.

Herzog

House

J. S.

Morgan

J. N.

(1991) Relation of work and retirement to health and well-being in older age. Psychology and Aging 6(2): 202.

23.

Hill

S. C.

Snell

A. F.

Sterns

H. L.

(2015) Career influences in bridge employment among retired police officers. The International Journal of Aging and Human Development 81(1–2): 101–119.

24.

Hobfoll

S. E.

(2002) Social and psychological resources and adaptation. Review of General Psychology 6(4): 307.

25.

Kalleberg

A. L.

(2009) Precarious work, insecure workers: Employment relations in transition. American Sociological Review 74(1): 1–22.

26.

Kanfer

(2015) Employment transitions in late adulthood. In: Finkelstein

Truxillo

Fraccaroli

Kanfer

(eds) Facing the challenges of a multi-age workforce: A use-inspired approach, New York, NY: Routledge, pp. 339–343.

27.

Kanfer

Chen

Pritchard

R. D.

(2012) Work motivation: Past, present and future, New York, NY: Routledge.

28.

Kelly

G. A.

Lazarus

(2015) Perceptions of successful aging: Intergenerational voices value well-being. The International Journal of Aging and Human Development 80(3): 233–247.

29.

Kooij

D. T.

De Lange

A. H.

Jansen

P. G.

Kanfer

Dikkers

J. S.

(2011) Age and work-related motives: Results of a meta-analysis. Journal of Organizational Behavior 32(2): 197–225.

30.

Koopmann

Wang

(2015) Resource-based dynamic perspective. In: Whitbourne

S. K.

(ed.) The encyclopedia of adulthood and aging, Hoboken, NJ: Wiley-Blackwell, pp. 1216–1220.

31.

Kreisberg

(2015) New careers for older workers, Great Barrington, MA: American Institute for Economic Research.

32.

Loi

J. L. P.

Shultz

K. S.

(2007) Why older adults seek employment: Differing motivations among subgroups. Journal of Applied Gerontology 26(3): 274–289.

33.

Morelock, J. C., McNamara, T. K., & James, J. B. (2017). Workability and requests for flexible work arrangements among older adults the role of a time and place management intervention. Journal of Applied Gerontology, 36(11), 1370–1392. doi:10.1177/0733464815624149.

34.

Morin

R. T.

Midlarsky

(2016) Social support, mastery, and psychological distress in black and white older adults. The International Journal of Aging and Human Development 82(2–3): 209–228.

35.

Raymo

J. M.

Warren

J. R.

Sweeney

M. M.

Hauser

R. M.

J.-H.

(2011) Precarious employment, bad jobs, labor unions, and early retirement. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 66(2): 249–259.

36.

Rhee

(2013) The retirement savings crisis, Washington, DC: National Institute on Retirement Security.

37.

Riumallo-Herl

Basu

Stuckler

Courtin

Avendano

(2014) Job loss, wealth and depression during the great recession in the USA and Europe. International Journal of Epidemiology 43(5): 1508–1517.

38.

Shultz

K. S.

(2003) Bridge employment: Work after retirement. In: Adams

G. A.

Beehr

T. A.

(eds) Retirement: Reasons, processes, and results, New York, NY: Springer, pp. 214–241.

39.

Shultz

K. S.

Taylor

M. A.

Morrison

R. F.

(2003) Work related attitudes of naval officers before and after retirement. The International Journal of Aging and Human Development 57(3): 259–274.

40.

Solberg

V. S.

Good

G. E.

Fischer

A. R.

Brown

S. D.

Nord

(1995) Career decision-making and career search activities: Relative effects of career search self-efficacy and human agency. Journal of Counseling Psychology 42(4): 448.

41.

Toossi

(2013) Labor force projections to 2022: The labor force participation rate continues to fall. Monthly Labor Review 136: 1.

42.

Wang, M., Olson, D. A., & Shultz, K. S. (2012). A resource-based dynamic perspective on mid and late careers. In Mid and late career issues: An integrative perspective (pp. 173–182). New York, NY: Routledge.

43.

Wang

Shultz

K. S.

(2010) Employee retirement: A review and recommendations for future investigation. Journal of Management 36(1): 172–206.

44.

Wise

D. A.

(2010) Facilitating longer working lives: International evidence on why and how. Demography 47(1): S131–S149.

45.

Wise

A. J.

Millward

L. J.

(2005) The experiences of voluntary career change in 30-somethings and implications for guidance. Career Development International 10(5): 400–417.

46.

Yoo

S. K.

Lee

(2017) Positive approaches to mid-life careers. In: Oades

L. G.

Steger

M. F.

Fave

A. D.

Passmore

(eds) The Wiley Blackwell handbook of the psychology of positivity and strengths-based approaches at work, Chichester, England: John Wiley & Sons, Ltd, pp. 366–388.

47.

Zenger

Brähler

Berth

Stöbel-Richter

(2011) Unemployment during working life and mental health of retirees: Results of a representative survey. Aging & Mental Health 15(2): 178–185.

48.

Zhang

(2011) Gender and marital status differences in depressive symptoms among elderly adults: The roles of family support and friend support. Aging & Mental Health 15(7): 844–854.