Life satisfaction across life course transitions

Content type
Research report
Published

September 2015

Researchers

Lixia Qu, David de Vaus

This paper explores how people’s life satisfaction changes across the major phases of life. It looks at how leaving home, making and breaking relationships, having children, “empty nesting”, retirement and widowhood affect how we feel about life.

Key messages

  • Life satisfaction declines from the age of 15 through to the mid-30s and is at its lowest between the mid-30s to the early 50s. However, it steadily improves until the late 60s and remains consistently high into the early 80s (near the highest level over a lifetime). Even when people are in their 80s, life satisfaction still remains higher than most of the earlier life periods.

  • Living with a partner increases and sustains life satisfaction for both men and women.

  • Having children increases life satisfaction for men and women, particularly in the year prior to child birth. However, there is sharp decline in life satisfaction for parents in the years after childbirth.

  • Relationship separation leads to a sharp decline in life satisfaction, but people who re-partner fare better than those who remain single. For those remaining single after a break up, women recover life satisfaction better than men.

  • Partnered empty nesters had higher life satisfaction than those who were single.

We all experience ups and downs at various points throughout life. This paper explores whether we experience predictable changes in life satisfaction over the life course—both as we get older and as we pass through a range of common life course transitions. The paper looks at how life satisfaction changes as we pass through various life transitions on the pathway from adolescence through young adulthood, middle age and into old age. It explores how leaving home, commencing a relationship, having children, separating, entering the “empty nest” phase, retirement and widowhood affect life satisfaction.

We concentrate on three main questions:

  • Do these life course transitions have an effect on life satisfaction, and if so, in what way?
  • Is any such effect of life course transitions greater for men or for women?
  • When a life course transition affects life satisfaction, is the change longlasting or relatively short-lived? For example, do people tend to “bounce back” within some years of experiencing transitions that erode life satisfaction, and if so, how long after the transition is this likely to become apparent? The same questions arise in relation to transitions that enhance life satisfaction.

Research into subjective wellbeing suggests that most people tend to view life in a positive light and are able to adapt to various circumstances (Cummins, 2000; Diener, Lucas, & Scollon, 2006). It is argued that, except in extreme circumstances, people have their own “normal” level of wellbeing, which increases or decreases in response to personally important changes and challenges, but eventually returns to their “normal” level. That is, the ups and downs of wellbeing will usually be transitory and are governed by some sort of homeostatic mechanism.

But people do not always bounce back—especially under severe circumstances (e.g., Cummins, 2000; Diener et al., 2006; Lucas, 2007). Diener et al. argued that individuals may have multiple “set points”, such that after experiencing adversity a person may bounce back, but not all the way, and thus they establish a new (but lower) baseline of life satisfaction. Cummins also maintained that extremes in adversity can defeat the homeostatic process.

Life satisfaction across the life course

Does life satisfaction change as people grow older? Is there a steady improvement as people move from adolescence through young adulthood, middle age and to old age? Or does life satisfaction decline as people move from the hopes and aspirations of youth through some of the harsher realities of adult life and the challenges of later life? Is there a predictable trajectory?

Figure 1 shows how life satisfaction varies with age from 15 years onwards. Using 12 years of data from the Household, Income and Labour Dynamics in Australia (HILDA) survey (see Box 1), Figure 1 shows that life satisfaction:

  • sharply declines from the age of 15 to the early 20s;
  • gradually declines from the early 20s to mid-30s;
  • being stable at a lower level from the mid-30s to the early 50s, the period of lowest life satisfaction;
  • begins to improve from the early 50s until the late 60s, both steadily and substantially;
  • remains steady and high from the late 60s to the early 80s, near the highest level over the life course; and
  • begins to decline a little from the early 80s, but nevertheless remains higher than that apparent for most earlier periods of life.

Figure 1: Mean ratings of life satisfaction, by age, 2001−12

Figure 1: Mean ratings of life satisfaction, by age, 2001−12. As described in text.

Notes: The life satisfaction rating of each person is included for each survey wave in which they participated; that is, if a person aged 20 in Wave 1 participated in all 12 waves, the rating of life satisfaction would be included for ages 20 to 32 years. Life satisfaction is assessed through the question: "All things considered, how satisfied are you with your life?" Response options range from 0 = completely dissatisfied to 10 = completely satisfied. 
Source: HILDA Waves 1−12

Box 1: The HILDA survey

The Household, Income and Labour Dynamics in Australia (HILDA) survey is a national household panel study that commenced in 2001 (see <www.melbourneinstitute.com/hilda/> for details). Wave 1 involved 7,682 private households nationwide and face-to-face interviews with 13,969 respondents aged 15 and over. The panel members are interviewed annually, and original respondents who had household members who turned 15 years and persons who became partners of original sample members were also interviewed in later waves. From Wave 9, the sample included new household members who had arrived in Australia for the first time after 2001. In Wave 11, the sample added an additional 2,153 households. Over the 12 waves, 2001−12, 26,953 respondents were interviewed in at least one wave, generating 165,169 person-wave records.

The HILDA project was initiated and is funded by the Australian Government Department of Social Services (DSS) and is managed by the Melbourne Institute of Applied Economic and Social Research. The findings and views based on these data are those of the authors and should not be attributed to either DSS or the Melbourne Institute.

However, there are some differences between the levels of life satisfaction of men and women at different ages. Figure 2 shows that:

  • in adolescence, males tend to be a little more satisfied than females;
  • from the age of 20 through to the mid-50s, women are, on average, more satisfied with their life than men; and
  • from the mid-60s onwards, men and women tend to be equally satisfied.

These changes in life satisfaction as people grow older partly reflect the different effects of maturation and aging. Another reason may be because age is linked to a range of life course transitions that can affect wellbeing. In the remainder of this paper, we explore whether major life events lead to improvements in or loss of life satisfaction, and whether any such effects persist in the longer term.

Figure 2: Mean ratings of life satisfaction, by age and gender, 2001−12

Figure 2: Mean ratings of life satisfaction, by age and gender, 2001−12. As described in text.

Notes: The life satisfaction rating of each person is included for each survey wave in which they participated; that is, if a person aged 20 in Wave 1 participated in all 12 waves, the rating of life satisfaction would be included for ages 20 to 32 years. Life satisfaction is assessed through the question: "All things considered, how satisfied are you with your life?" Response options range from 0 = completely dissatisfied to 10 = completely satisfied. 
Source: HILDA Waves 1−12

The effect of life events

Research suggests different patterns of adaption to major life events and transitions. Qu, Baxter, Weston, Moloney, and Hayes (2012) compared various wellbeing indicators before and after a range of life events. They found that certain life events, such as major financial crises and relationship breakdown, were associated with declines in a sense of wellbeing, and that those who experienced such life events also tended to have lower wellbeing prior to the life event. However, events such as retirement were not associated with a decline in a sense of wellbeing.

Lucas (2007) tracked the effect of six life events (marriage, widowhood, divorce, unemployment, disability and severe disability) on life satisfaction. Using longitudinal data from Britain and Germany, he looked at life satisfaction from five years before the life event and for seven years afterwards and found that both the effect and the adaption process varied with different life events. For example, while the life satisfaction of both those who became widowed and those who experienced divorce showed some recovery after the event, the extent of recovery was greater for widows than divorcees. In contrast, those who became severely disabled experienced a loss of life satisfaction from which they did not recover.

There are various ways in which the effect of a life course event on life satisfaction might be studied.

One approach is to use cross-sectional data and simply compare life satisfaction between those who have and have not experienced the event. Thus the life satisfaction of divorced people might be compared with that of non-divorced people, or the life satisfaction of retired people compared with non-retired people. The obvious problem with this approach is that those who have experienced the life event will differ in many other respects from those who have not experienced the event. While some of these factors can be controlled statistically, we cannot be certain that the remaining differences in life satisfaction are due to the life event.

A better approach is to track individuals who experience the event and their life satisfaction over time. This is the approach used in this study. A change in life satisfaction following the event may indicate that this was influenced by the event. But the change might also be part of a longer term change in life satisfaction that started well before the event. Thus, a tracking study needs to be able to look at any changes in life satisfaction that were taking place prior to the event before it can ascertain the effect of the life event itself. In this paper, we are able to see whether or not any changes in life satisfaction pre-date the actual life course transition.

An issue for tracking studies is how long to track a person both before and after a life event. In this study, we used data from the first 12 waves of the HILDA survey to track people from up to six years before the life event to up to six years following the event (a similar approach to that adopted by Lucas, 2007). Using this design, we could assess both the short- and longer term effects of the life event and see whether any changes were initiated by the life event or whether they were simply a continuation of changes that had commenced well before.

The analysis focuses on changes in life satisfaction, with seven key milestones over the life course:

  • leaving the parental home;
  • moving in with a partner;
  • childbirth;
  • separation from a live-in relationship;
  • becoming empty nesters;
  • retirement; and
  • widowhood.

See Box 2 for a detailed description of each of these life transitions and how they were measured.

Box 2: Life course transition measures

Leaving the parental home is based on data concerning relationships between each person in the household. Respondents were only considered to be staying in the parental home if they were either dependent or non-dependent children of a household member. If they subsequently moved to a household where neither of their parents lived, then they were considered to have left their parental home.

The measure moving in with a partner focused on those who were single in one year and partnered in all six subsequent years. Samples of post-transition time points only focused on those who had been with the same partner throughout the period examined.

Identification of childbirth is based on changes in the total number of children in each wave; that is, the increase in the number of children from none to one, one to two, two to three and so on.

For separation from a live-in relationship, no differentiation is made in this analysis between those who re-partnered and those who remained single (except where this is the specific focus). The spell of separation ends if there is separation from a new partner.

Becoming empty nesters represents the period when single or partnered parents who had been living with at least one dependent or non-dependent child are no longer doing so. The empty-nest state is considered to continue for as long as no such children are recorded as living with these parents.

Retirement was identified based on whether respondents either had retired completely from the workforce in Waves 1–2, 5–6, 8−10 and 12; or considered themselves to be completely retired from the paid workforce, partly retired or not retired at all in Waves 3, 7 and 11. Retirement was not asked about in Wave 4. In this paper, retirement refers to complete retirement from the workforce. If respondents did not respond to either option in a specific wave and were in the labour force (or marginally attached to the workforce), they were classified as not retired. A small number of respondents indicated that they had completely retired and also that they were still in the labour force or marginally attached to the labour force. The data for these respondents were omitted from the analyses.

Widowhood includes respondents who were married or cohabiting, but whose partner died in a subsequent wave.

The analysis is based on “unbalanced samples” for all the transitions, with some participants responding over a longer timespan and some over a shorter timespan. Therefore, sample sizes vary at each time point of each transition.

Respondents’ life satisfaction was assessed through the question: “All things considered, how satisfied are you with your life?” Response options range from 0 = completely dissatisfied to 10 = completely satisfied. Mean ratings of respondents aged 15 years and over ranged from 7.8 to 7.9 across 12 waves, with 65–69% providing ratings of 8 or above (indicating high life satisfaction). Over all waves, 2.7% of ratings were in the 0–4 range, 29% were in the 5–7 range and 68% were in the 8−10 range.

Leaving the parental home

Leaving home does not lead to any immediate change in life satisfaction, but it may help arrest a decline in life satisfaction, especially for women. Figure 3 shows that:

  • on average, there was no change in life satisfaction from the year before leaving home to the year immediately afterwards for either men or women;
  • however, there was a decline in average life satisfaction over the extended period, with this decline beginning well before leaving home, so the decline cannot be attributed to leaving home; and
  • gender differences were apparent four years after leaving home, with women being more satisfied than men.

Figure 3: Mean ratings of life satisfaction before and after leaving the parental home, men and women, 2001−12

Figure 3: Mean ratings of life satisfaction before and after leaving the parental home, men and women, 2001−12. As described in text.

Notes: No. of transitions—men: 1,398; women: 1,361. Sample sizes across time points—men: 313−1,398; women: 253−1,361 (the further from the event point, the smaller the sample). The vertical line indicates the event. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity. 
Source: HILDA Waves 1−12

The experience of leaving home will not be the same for all people. Figure 4 shows how the effect of leaving home differs for men and women depending on whether they leave home and remain single or whether they leave home to live with a partner.

Figure 4 depicts the changes in average life satisfaction for those who were living with a partner (panel A) and those who remained single (panel B) in all waves after leaving their parental home. Those who remained single left home for various reasons, such as moving to go to university or for work, because family relationships were stressed, or to achieve independence.

The two groups show quite different trajectories in life satisfaction both before and after leaving home.

  • Even six years prior to leaving home, those who remained single (especially women) tended to report lower levels of satisfaction than those who moved out to live with a partner.
  • Those who left home to live with a partner showed a relatively stable level of life satisfaction in the years before leaving home, whereas those who left home but remained single had already experienced a sharp decline in life satisfaction in the years before leaving home.
  • Prior to leaving home, the level of life satisfaction of those who remained single dropped more sharply for females than for males.
  • Further analysis (results not shown here) suggests that those who left home for reasons other than living with a partner were less satisfied with their relationship with their parents compared with those who left home to live with a partner. This pattern was consistent both before and after leaving home for men and women. Thus, it appears that poor relationships with their parents led some young adults to leave home.
  • After leaving home, those who partnered had higher life satisfaction than those who remained single.
  • After leaving home, women who partnered showed a modest increase in life satisfaction, while men experienced a decline after being partnered for two years.
  • After leaving home and remaining single, both men and women continued to experience further declines in life satisfaction, but less so than while they were still living at home, with life satisfaction tending to fluctuate somewhat after two years of living away from home. In general, however, leaving home appeared to slow the rate at which life satisfaction declined for those who remained single.

Figure 4: Mean ratings of life satisfaction before and after leaving the parental home, by whether living with a partner or remaining single, men and women, 2001−12

Figure 4: Mean ratings of life satisfaction before and after leaving the parental home, by whether living with a partner or remaining single, men and women, 2001−12. As described in text.

Notes: No. of transitions—men: 370 (partner), 684 (single); women: 419 (partner), 594 (single). Sample sizes across time points—all the groups: 94–684 (the further from the transition point, the smaller the sample). Excludes those who had been single in one or more waves and lived with a partner in other waves after leaving the parental home. The vertical line indicates the event. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity. 
Source: HILDA Waves 1−12

Moving in with a partner

Figure 5 shows average trends in men’s and women’s levels of satisfaction during the six years before and after they entered a live-in relationship with a partner, regardless of their living arrangements before this event (e.g., they may have been living with parents, other family or friends, boarding or living alone).

Beginning to live with a partner was linked with a marked and then sustained increase in life satisfaction for both men and women. Figure 5 shows that:

  • in the six years prior to partnering, life satisfaction was either stable or declining marginally;
  • the year following partnering reflected a sharp increase in life satisfaction, with the rise appearing to start a year before partnering for men; and
  • this increase in satisfaction was sustained over the following six years.

Figure 5: Mean ratings of life satisfaction before and after moving in with a partner, men and women, 2001−12

Figure 5: Mean ratings of life satisfaction before and after moving in with a partner, men and women, 2001−12. As described in text.

Notes: No. of transitions—men: 1,699, women: 1,826. Sample sizes across time points—men: 430−1,699, women: 446−1,826 (the further from the event point, the smaller the sample). The vertical line indicates the event. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity. 
Source: HILDA Waves 1−12

Does the effect of partnering depend on whether a person is marrying or cohabiting, and does it depend on whether a couple live together or not before marrying? This issue is addressed in Figure 6.

Panels A and B in Figure 6 show the apparent effects on life satisfaction of commencing a new live-in relationship, according to whether this new arrangement occurred after marriage (here called a “direct marriage”—the couples had not lived together beforehand) or after cohabiting (a de facto relationship). The data show that:

  • While couples married directly, life satisfaction increased before the marriage, the increase occurred earlier for women than men.
  • The effects associated with marriage were greater for women than for men.
  • Right before commencing cohabiting with a partner, life satisfaction tended to increase, especially for women.
  • After marrying directly, on the other hand, the level of increase in life satisfaction tended to be greater for men than women. This is not surprising, given that women’s life satisfaction increased prior to their marriage.

Panel C in Figure 6 shows levels of life satisfaction of those who lived together before marrying. This arrangement is here referred to as “indirect marriage”. The following trends were apparent for indirect marriage:

  • Life satisfaction rose before marrying for both genders, especially women.
  • For women, marrying after living together did not lead to an increase in life satisfaction, and showed slightly higher level of life satisfaction than those who married directly.
  • Men who married after cohabiting displayed a short-term increase in life satisfaction after marrying and then steadied before it dropped to pre-marriage level.

Overall, live-in relationships appeared to have a greater effect on women than men, though the prospect of marrying indirectly appears to provide men with a boost in satisfaction that remains higher after marriage for up to four years. This boost for women is prior to marriage and higher than those who marry directly.

Figure 6: Mean ratings of life satisfaction before and after marriage or cohabitation, never married men and women, 2001−12

Figure 6: Mean ratings of life satisfaction before and after marriage or cohabitation, never married men and women, 2001−12. As described in text.

Notes: No. of transitions—men: 114 (direct marriage), 828 (cohabitation), 214 (indirect marriage); women: 149 (direct marriage), 941 (cohabitation), 223 (indirect marriage). The vertical lines in panels A and C indicate the event of marriage, while the vertical line in panel B marks the start of cohabitation. For the group married after cohabitation, the time before marriage refers to the period of cohabitation. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity. 
Source: HILDA Waves 1−12

Childbirth

The birth of a child was linked to a rise in life satisfaction prior to the birth and a decline in life satisfaction after the birth for both men and women (Figure 7). However, there were some important differences in the way in which the birth of a child affected the life satisfaction of mothers and fathers:

  • While life satisfaction increased for both genders in the year prior to having a child and was highest in the year immediately prior to having a child, the pattern was stronger for women than men.
  • Following the birth of a child, the life satisfaction of women declined a little in the first year and then sharply in the second year.
  • Among men, there was little noticeable change in life satisfaction in the year immediately following the birth.
  • In all except the sixth year after the birth of a child, the life satisfaction of men and women declined steadily.

Figure 7: Mean ratings of life satisfaction before and after childbirth, men and women, 2001−12

Figure 7: Mean ratings of life satisfaction before and after childbirth, men and women, 2001−12. As described in text.

Notes: No. of transitions—men: 2,465, women: 2,694. Sample sizes across time points—men: 513–2,465, women: 484–2,694 (the further from the event point, the smaller the sample). The vertical line indicates the event. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity. 
Source: HILDA Waves 1−12

The results above refer to the birth of any new child. Does the birth of a first child have more influence on life satisfaction than subsequent children?

Figure 8 shows the extent to which life satisfaction varies for mothers and fathers when they have their first, second and subsequent children. The direction of the trends is similar for fathers and mothers:

  • The anticipation of the first child was linked to a sharp increase in life satisfaction, especially for women. In the first year after the birth of the first child, life satisfaction of both mothers and fathers remained high but thereafter dropped sharply.
  • The birth of a second child shows a similar, but more muted, pattern as for the first child—increasing satisfaction in the years before the second child, and then satisfaction declining steadily in the years following the birth.
  • Third and subsequent children produce a similar but even more muted change in life satisfaction—a modest increase in life satisfaction before the birth and a modest decline after the birth.
  • It is noticeable that in the fifth year after the birth of both the first and second child, the decline in life satisfaction is reversed and shows a small improvement. Time will tell if this improvement is associated with children commencing school or is simply a fluctuation.
  • Overall, parents anticipate each birth with an increase in life satisfaction and appear to respond to some of the demands of childrearing with a decline in life satisfaction. With more experience, the anticipatory effects become less marked and the subsequent readjustments less severe.
  • Although the overall trends are similar for men and women, the anticipatory effect and post-birth fall in life satisfaction is greater for women than men. This pattern is consistent for all children, no matter the birth order. The greater fall in life satisfaction among mothers is partly because of their heightened increase in life satisfaction preceding the birth.

Figure 8: Mean ratings of life satisfaction before and after childbirth, by child’s birth order, men and women, 2001−12

Figure 8: Mean ratings of life satisfaction before and after childbirth, by child’s birth order, men and women, 2001−12. As described in text.

Notes: No. of transitions—men: 946 (1st child), 836 (2nd child), 688 (3rd+ child); women: 1,066 (1st child), 940 (2nd child), 689 (3rd+child). Sample sizes across time points—all groups: 61−1,066 (the further from the event point, the smaller the sample). The vertical line indicates the event. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity. 
Source: HILDA Waves 1−12

Separation from a live-in relationship

Relationship separation was linked with a sharp decline in life satisfaction among both men and women (Figure 9).

  • The decline in life satisfaction began well before separation for women, while this pattern was less apparent for men. So while the event of separation accentuates the decline in life satisfaction, it simply reflects a longer term decline, especially for women.
  • The life satisfaction of women declined more sharply in the years prior to separation than it did for men. While the fall in life satisfaction right after the separation was greater for men.
  • Following separation, the life satisfaction of both men and women improved steadily for the next six years, but only recovered to the level that was apparent about two years prior to separation.
  • In other words, the level of life satisfaction six years following separation was considerably lower than it was six years prior to separation, though this may well continue to increase in the years ahead.

Figure 9: Mean ratings of life satisfaction before and after separation from a live-in relationship, men and women, 2001−12

Figure 9: Mean ratings of life satisfaction before and after separation from a live-in relationship, men and women, 2001−1. As described in text.

Notes: No. of transitions—men: 1,404, women: 1,671. Sample sizes across time points—men: 297−1,404, women: 394−1,671 (the further from the event point, the smaller the sample). The vertical line indicates the event. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity. 
Source: HILDA Waves 1−12

Following separation, people can take different pathways. Some separated people remain single indefinitely, while others re-partner within the first few years of separation. While research suggests that the effects of relationship separation on personal wellbeing can be longlasting, it appears that people who re-partner recover better (de Vaus, Gray, Qu, & Stanton, 2010; Lucas, 2005).

What do the longitudinal data reveal about how separation followed by re-partnering or remaining single affects wellbeing?

Figure 10 sheds further light on men’s and women’s life satisfaction not just after separation, but in the years before for those who eventually re-partnered or who remained single. Panel A of Figure 10 shows those who separated but remained single for the full six years after separation. Panel B shows those who separated but re-partnered at some point within the six years following separation.

  • Regardless of re-partnering status, the change in life satisfaction before and after separation was consistent with the broad pattern described earlier—life satisfaction started to fall two to three years before separation, reaching the lowest point right after separation, then began to recover.
  • For both men and women, the life satisfaction of those who eventually re-partnered recovered to much higher levels than their same-gender counterparts who remained single after separation.
  • Both men and women who remained single were still less satisfied six years after separation compared with the period before separation.
  • The low point in life satisfaction was in the year immediately following separation, in which period very few (5%) had re-partnered. But for both men and women, the low point was much lower for those who remained single than for their counterparts who eventually re-partnered. Men who remained single had the lowest level of life satisfaction of all the four groups.
  • Following separation, the life satisfaction of women who remained single recovered better than was apparent for separated men who remained single, but these women also seemed more satisfied than the men well before separation occurred.
  • Gender differences are more pronounced for those who remained single than for those who re-partnered.

Figure 10: Mean ratings of life satisfaction before and after separation, by re-partnering status, men and women, 2001−12

Figure 10: Mean ratings of life satisfaction before and after separation, by re-partnering status, men and women, 2001−12. As described in text.

Notes: No. of transitions—men: 873 (remained single), 533 (re-partnered); women: 1,152 (remained single), 524 (re-partnered). Re-partnered respondents are those whose re-partnering took place at some point within the six years following separation. Sample sizes across time points—all groups: 96−1,152 (the further from the event point, the smaller the sample). The vertical line indicates the event. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity. 
Source: HILDA Waves 1−12

Becoming empty nesters

Any effects from seeing the only or last child leaving the family home (and thus becoming empty nesters) is relatively modest.1  Figure 11 indicates that:

  • there was a small increase in life satisfaction in the years following the commencement of the empty nest stage; and
  • this improvement did not seem to be due to the empty nest, as gradual improvement was evident in most of the six years prior to the nest emptying—maybe reflecting declining parental responsibility and stresses prior to the final child departing.

Figure 11: Mean ratings of life satisfaction before and after becoming empty nesters, men and women, 2001−12

Figure 11: Mean ratings of life satisfaction before and after becoming empty nesters, men and women, 2001−12. As described in text.

Notes: No. of transitions—men: 938, women: 1,175. Sample sizes across time points—men: 292–938, women: 357−1,175 (the further from the event point, the smaller the sample). The vertical line indicates the event. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity 
Source: HILDA Waves 1−12

Some empty nests are emptier than others—the empty nest will be emptier for single parents than for couples. Does being a single or partnered parent make any difference to life satisfaction at the empty nest phase?

Figure 12 indicates that the level of life satisfaction of couple parents was considerably higher than that of single parents both before and after the nest empties:

  • Among single parents, the satisfaction of mothers improved when the only or last child left home, while that of fathers remained essentially unchanged.
  • While single parents were less satisfied with life compared with couple parents, single fathers were the least satisfied of the four groups of parents, and they were the only group that showed no sign of an increase in life satisfaction over time. (It should be noted that the marked fluctuation in satisfaction of these fathers may be a function of their small sample sizes across the time points.)
  • Couple parents appeared to be on the path of a slow increase in life satisfaction in the years with children still at home and this pattern continued after children left home.

Figure 12: Mean ratings of life satisfaction before and after becoming empty nesters, by family type, men and women, 2001−12

Figure 12: Mean ratings of life satisfaction before and after becoming empty nesters, by family type, men and women, 2001−12. As described in text.

Notes: No. of transitions—men: 150 (single parents), 788 (couples); women: 353 (single parents), 822 (couples). Sample sizes across time points—all groups: 50–822 (the further from the event point, the smaller the sample). The vertical line indicates the event. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity. 
Source: HILDA Waves 1−12

Retirement

In Figure 13 there is some evidence of an improvement in life satisfaction following retirement:

  • In the five years prior to retirement, life satisfaction is very stable.
  • In the year immediately following retirement, there is no change in life satisfaction for men and a marginal though non-significant change for women.
  • Thereafter, the life satisfaction of both men and women mostly improves.

Figure 13: Mean ratings of life satisfaction before and after retirement, men and women, 2001−12

Figure 13: Mean ratings of life satisfaction before and after retirement, men and women, 2001−12. As described in text.

Notes: No. of transitions—men: 802, women: 872. Sample sizes across time points—men: 219–802, women: 225–872 (the further from the event point, the smaller the sample). The vertical line indicates the event. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity. 
Source: HILDA Waves 1−12

People take different pathways to retirement. For some people, retirement is a process that involves moving from full-time to part-time work before retiring completely. For others, retirement follows immediately from full-time work. There are also those who move from being not employed (whether due to illness, unemployment or not being in the labour force) to defining themselves as being retired. For these people, “retirement” involves a change in self-definition rather than in activities.

Does the pathway to retirement make any difference to life satisfaction?

Figure 14 shows men’s and women’s life satisfaction by pathways to retirement.

  • Across all years, those who indicated that they had retired following being inactive in the labour force had lower life satisfaction upon retirement than those who retired from either full-time or part-time work. That is, their low post-retirement levels of satisfaction relative to the other groups was not a function of their change in self-definition from being not employed to retired.
  • Nevertheless, the life satisfaction of men and women who retired following no employment activity improved.
  • The life satisfaction of men who retired directly from full-time work improved immediately after retirement and continued to improve into the second year. Despite fluctuations, their life satisfaction remained at a relatively high level in the subsequent years.
  • Men and women who retired from part-time work appeared to lose some of their satisfaction with life initially and then recovered.

Figure 14: Mean ratings of life satisfaction before and after retirement, by pathways to retirement, men and women, 2001−12

Figure 14: Mean ratings of life satisfaction before and after retirement, by pathways to retirement, men and women, 2001−12. As described in text.

Notes: "Pathway to retirement" refers to the individual’s employment status right before retirement (i.e., time point −1). No. of transitions to retirement—men: 238 (full-time), 264 (part-time), 299 (not employed); women: 385 (part-time) 363 (not employed). The sample sizes for women retiring from full-time employment are small and thus not shown. Sample sizes across time points—all groups: 67–363 (the further from the event point, the smaller the sample). The vertical line indicates the event. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity. 
Source: HILDA Waves 1−12

Widowhood

Figure 15 shows that life satisfaction hits a low point in the year immediately following the death of a partner, but:

  • life satisfaction begins declining sharply for both men and women in the two or three years prior to the partner’s death;
  • following the low point in the first year after widowhood, life satisfaction recovers rapidly and sharply, but not to the levels of six years prior to their partner’s death; and
  • the recovery in life satisfaction is more rapid for women than men, but by six years after widowhood, men and women have attained the same level of life satisfaction.

Figure 15: Mean ratings of life satisfaction before and after widowhood, men and women, 2001−12

Figure 15: Mean ratings of life satisfaction before and after widowhood, men and women, 2001−12. As described in text.

Note: No. of transitions—men: 138, women: 322. Sample sizes across time points—men: 32−138, women: 135—322 (the further from the event point, the smaller the sample). The vertical line indicates the event. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity. 
Source: HILDA Waves 1−12

Gender differences

Throughout this paper we have noted gender differences in the way in which life satisfaction changes in the six years before and after particular life events.

The overall life satisfaction trajectories of men and women were shown in Figure 2, while Figure 16 shows trends in the life satisfaction of men and women across all seven life transitions discussed above. Broadly speaking, men and women respond in similar ways to the life course transitions examined here. The life satisfaction trajectories in Figures 2 and 16 are similar for both men and women. But similar is not identical, for men and women tended to differ in the degree to which these life events appeared to affect their life satisfaction and the speed of the recovery (i.e., returning to a “normal” level).

Figure 16: Mean ratings of life satisfaction before and after each life course transitions, men and women, 2001−12

Figure 16: Mean ratings of life satisfaction before and after each life course transitions, men and women, 2001−12. As described in text.

Notes: No. of transitions—men: 803–2,470, women: 872–2,695 (except widowhood—men: 139, women: 326). Sample sizes across time points—all groups: 250 or more, except widowhood—men: 52 (before), 60 (after); women: 135 (before), 150 (after) (the further from the event point, the smaller the sample). The vertical line indicates the event. The interval between years −1 and 1 is about 12 months on average but it is marked as lying between year −1 and year 1 for simplicity. 
Source: HILDA Waves 1−12

Leaving the parental home was preceded by a period of declining life satisfaction for both men and women and was followed by a pause in this decline. However, for men a slow decline in life satisfaction resumed a year or two later before stabilising again, while for women life satisfaction improved a little.

Moving in with a partner was linked with improved live satisfaction for both men and women, but a little more for women than for men.

Childbirth was linked with increased life satisfaction for both men and women, but more sharply for women. A year or so after the birth, life satisfaction declined among both men and women, but more sharply for women again—partly because it had risen more sharply just prior to the birth.

Changes in life satisfaction associated with separation from a live-in relationship followed a similar general pathway for both men and women. Life satisfaction declined in the two or three years prior to separation, but more sharply for women, given that women were more satisfied with life well before separation. The low point for both men and women was in the year following separation, after which the life satisfaction of both women and men steadily improved.

The pre- and post-empty nest phases were similar for men and women. There was a slow improvement in life satisfaction for both men and women in the few years prior to becoming empty nesters and this improvement continued in subsequent years.

Retirement had similar effects on life satisfaction for both men and women. Life satisfaction was stable in the five years before retiring, and remained fairly steady in the year immediately after retirement. Thereafter life satisfaction improved a little—perhaps slightly more for men than women.

Life satisfaction also followed a similar trajectory for men and women following widowhood. There was a steady decline in life satisfaction in the years before their partner’s death with the low point being immediately after the death. Thereafter, life satisfaction improved for both men and women but more rapidly for women. Six years after widowhood, both men and women had recovered to the same level.

The more fine-grained analyses relating to these transitions outlined in the previous figures indicated some more varied responses to the various life transitions, depending on the context.

Summary

People are generally satisfied with their life, regardless of the transitions that commonly occur across the life course. Both men and women have mean life satisfaction ratings above 7 on a 0−10 scale before, during and after the transitions examined in this paper.

But this is not to say that men and women experience the same level of life satisfaction or that the levels of life satisfaction are not affected by the various transitions experienced over the life course. When tracked over a 12-year period we see changes in life satisfaction that sometimes precede and often follow various life course transition points. The discussion in this paper has pointed to the ways in which life satisfaction changes over the life course. The headline points can be summarised as follows:

  • The life transitions of leaving the parental home and becoming an empty nester appeared to have little or no overall effect on life satisfaction. Retirement appeared to be linked with a slight improvement in life satisfaction.
  • Commencing to live with a partner and having a new child saw a rise in life satisfaction. However, life satisfaction following the birth of a child initially rose but then declined in the first five years.
  • Life satisfaction reached a low point immediately following relationship separation and becoming widowed, but improved thereafter. For both separation and widowhood, life satisfaction started to decline at least two or three years before the event took place. Despite recovering somewhat over the following six years, the level of life satisfaction did not return to the initial level that was evident six years before the events.

This article also highlights that the link between life satisfaction and life course transitions depends partly on the context of the transition. That is, the circumstances of people as they go through a transition can affect the way in which the transition was associated with life satisfaction.

  • When people left their parental home to live with a partner, life satisfaction after leaving home was higher than for those who left their parental home but remained single. This was especially the case among women.
  • Regardless of whether they married or cohabited, living with a partner was associated with higher levels of life satisfaction. Those who continued to cohabit for the subsequent six years appeared to be less satisfied with life than those who married (directly or indirectly) both before and after the transition.
  • The post-birth decline in parental life satisfaction was greatest for the first birth and least apparent for the third or subsequent children.
  • After relationship separation, those who re-partnered fared better than those who remained single.
  • Partnered empty nesters experienced higher life satisfaction than single-parent empty nesters. Single mothers who became empty nesters had the greatest increase in life satisfaction, but did not attain the level of life satisfaction apparent for partnered mothers and fathers who became empty nesters.

References

  • Cummins, R. A. (2000). Objective and subjective quality of life: An interactive model. Social Indicators Research, 52, 52–72.
  • de Vaus, D., Gray, M., Qu, L., & Stanton, D. (2010). Effect of relationship breakdown on income and social exclusion. In R. Sainsbury & P. Saunders (Eds.), Social security, poverty and social exclusion in rich and poorer countries. Portland, OR: Intersentia.
  • Diener, E., Lucas, R. E., & Scollon, C. N. (2006). Beyond the hedonic treadmill: Revising the adaptation theory of well-being. American Psychologist, 61(4), 305–314.
  • Lucas, R. E. (2005). Time does not heal all wounds: A longitudinal study of reaction and adaptation to divorce. Psychological Science, 12(6), 945–950.
  • Lucas, R. E. (2007). Adaptation and the set-point model of subjective well-being: Does happiness change after major life events? Current Directions in Psychological Science, 16(2), 75–79.
  • Qu, L., Baxter, J., Weston, R., Moloney, L., & Hayes, A. (2012). Family-related life events: Insights from two Australian longitudinal studies. Melbourne: Australian Institute of Family Studies.

1 Empty nesters include parents whose grown-up children left home again after returning home.

Acknowledgements

Dr Lixia Qu is a Senior Research Fellow at the Australian Institute of Family Studies. At the time of writing, Professor David de Vaus was a Senior Research Fellow at the Institute. He is currently Emeritus Professor at Queensland and La Trobe universities.

Cover © istockphoto/njgphoto

Citation

Qu L., & de Vaus, D. (2015). Life satisfaction across life course transitions (Australian Family Trends No. 8). Melbourne: Australian Institute of Family Studies.

ISBN

978-1-76016-039-5

Share