Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

The transition to adulthood and family life in Western industrialized countries has changed dramatically over the past 30 years. There is a growing body of literature describing the sorts of changes that characterize this modification of the early life course. These changes include a growing diversification in the temporal sequence of events and a delay in experience of many of them (Corijn and Klijzing 2001; Lloyd 2005; Osgood et al. 2005; Settersten et al. 2005; Shanahan 2000). Australia has now experienced many of the changes in family formation and the early life course experienced in other countries (van de Kaa 1987; Liefbroer 1991; Keirnan 2001; Gauthier 2007). We see greater numbers of young people accessing post-secondary education and there is much greater emphasis on formal qualifications now than in the past. We have also seen a rise in cohabitation, a delay in marriage, and an increase in childbearing outside of marriage (McDonald 2003).

Chapter 2 proposed two theoretical frameworks to understand these changes. First, second demographic transition theory suggests that events in early adulthood will change as a result of ideational change and the need for individuals to realize their own family formation aspirations. Second, the globalization framework suggests that family formation decisions are made in a situation where institutions limit and constrain the choices of individuals so that the society can effectively globalize. This chapter takes the approach that both ideational shifts and institutional constraints impact the decisions made by individuals in the early life course.

It is important to consider that social and economic movements, such as globalization and modernization, occur in cross-sectional time and individual birth cohorts will experience the effect of social change differently at different points in the life course. To exemplify the impact of social movements on birth cohorts we can compare the expected experience of two cohorts—those born from 1955 to 1964 and those born from 1975 to 1984. One of these cohorts falls into the popularly termed generation of baby boomers, whereas the younger cohort forms part of generation Y. Those born in the 1955–1964 birth cohort will have reached adolescence during the period of economic modernization. We would expect that their early life course behaviour would be impacted by changing economic and social conditions such that they would have experienced free tertiary education and high levels of employment as they entered adulthood (McDonald and Evans 2003). This is a period of changing ideology surrounding family formation (van de Kaa 1987) and while many of this cohort would not have experienced different family formation behaviours, cohabitation, delayed parenthood and smaller families would become part of the social dialogue during their early adulthood. They will also be the first cohort to access modern contraceptives such as the contraceptive pill.

On the other hand, people born in the 1975–1984 cohort entered adolescence during the late 1980s and early 1990s during the period of globalization. They are a cohort who witnessed enormous financial collapses and a recession. Their early adulthood is characterized by risk and uncertainty and their response is to delay life events such as leaving school and family formation (Beck 1992; Mills et al. 2005; Mills and Blossfeld, Chap. 2, this volume). Delayed school leaving equips individuals with the skills to enter a globalized labour force. Delayed family formation follows from increased investments in education and skill building, and is easier to manage within a globalized society. Unlike the older cohort, this cohort have witnessed the result of changing family norms: divorce, cohabitation and births outside marriage are commonplace.

As well as impacting directly on cohorts as they live through social changes there is also the intergenerational impact of social and economic change. As cohorts experience social change their parents (or children) are also being affected by the same changes but at a different stage of the life course. A young person considering the shape and nature of their future career is likely to be influenced by the experience of their parents financial or employment instability in periods of recession. Similarly, the attitudes of older people to newer social phenomenon, such as childbearing within cohabiting unions, may be moderated by the experience and behavior of their children (Evans and Gray 2005).

This chapter will describe the process of leaving home in Australia for cohorts born between 1945 and 1985. During the 1980s the age of leaving the parental home was increasing in many developed countries after a period of decline (Billari et al. 2001; Molgat 2002; Young 1987; Corijin and Klijzing 2001). A recent Australian study has found that leaving home still occurs during early adulthood and has not been delayed to the same extent as other demographic or lifecourse events (Kilmartin 2000). Leaving home is not an event that occurs in isolation but is influenced by education, employment and relationship formation at both the macro and individual levels. In order to contextualize leaving home, this chapter will explore these three additional factors in combination with leaving home.

Data and Method

The chapter uses data from four waves of the Negotiating the Life Course project. The total sample size is 4,321 respondents which includes the 2,231 original respondents from Wave 1 as well as the 2,000 new respondents introduced in Wave 4. A person-period data set was constructed, so that each individual has multiple records of data representing their experience at every age. For each of the outcomes of interest, leaving home, leaving full-time education and entering the labour market, individuals are followed until the outcome of interest occurred. If they did not experience the event they were followed to the age of the last interview or up to age 35 – whichever occurred first. For leaving home and entering the labour market the initial starting point for observing individuals was set at 15 years of age, while for leaving full-time education the initial starting point was set at 14. Throughout the chapter, analysis is conducted separately for males and females to account for the possible interaction effects between gender and the other variables, such as age, on the outcomes of interest. The following sections describe the measurement of the main events covered in this chapter; leaving home, exiting education and entering the labour market.

Age at Leaving Home and Reason for Leaving Home

The age the respondents left home was identified using both retrospective information from the first interview as well as prospective information at subsequent waves. At the first interviewFootnote 1 respondents were asked if they were still living in their parent’s home. If they were not living with their parents they were asked how old they were when they last left their parent’s home. A broad assumption was made that this was the age at which individuals first left their home, although in reality from the given information it is not possible to tell whether the respondent had left their parental home once, or whether they had left and returned multiple times. In the former case the age at which the respondent last left their parents would be equal to the age of first home leaving. In the latter case however the age of first home leaving is unknown.

If the respondents were living at home in the first wave they were asked if they have ever lived away, and if yes, at what age they last returned to live with their parents. For individuals in this situation the age at which they first left home is unknown and is therefore set to missing. If respondents were living at home in the first wave, and had never lived away from home, they were tracked over the subsequent waves. At each subsequent wave respondents who had previously lived at home were asked if they had now moved out, if they had moved out since the previous interview but had returned home, or if they had remained with their parents for the whole time since the previous interview. Those who were now living away from their parents, or who had done so at least once since the previous interview, were asked at what age they left the parental home. This is taken as the age of first leaving home.

Two separate pathways out of the parental home were identified, leaving home to enter a relationship or leaving home for another reason. Leaving home to enter a relationship was further sub-divided into leaving home to enter a cohabiting union or leaving home to enter a marriage. The distinction between the two main pathways was made by comparing the date of home leaving with the date at which the first live-in relationship (cohabitation or marriage) occurred. The date at which the first relationship occurred was known down to both the month and year in most cases. The date of leaving home was not given in as much detail, and was approximated by the date of birth plus the age of first home leaving converted into months. Due to the discrepancy in the level of detail between these two dates a 6 month window was used so the reason for leaving home was classified as being due to a relationship if a relationship started up to and including 6 months before or after the date of home leaving. If the start of a co-residential relationship did not coincide with leaving home, the respondent was classified as leaving home for other reasons.

Age of First Exit from Full-Time Education

The age at which the respondents’ first exited full-time education was also identified using retrospective and prospective information. The age was derived by combining information from the study calendar and also from the questions on highest level of schooling and post-school education. The study calendar collects information on individuals’ study history at every year from the year they turned 15 up to the current year. For each of the relevant years individuals could note if they were studying full-time, studying part-time or not studying at all. The age at which the first exit from full-time study occurred is taken to be the age at which the person was in full-time education for the last time before experiencing a break in education. If respondents indicated on the study calendar that they were not in full-time education at the age of 15, but that they had completed secondary school they were assumed to be in full-time education up to age 18. If they had incomplete secondary schooling and were not in school at age 15 they were assumed to have been in full-time education up to and including age 14.

Age of First Entry into the Labour Market

The age at which the respondents first entered the labour market (full-time labour market, part-time labour market and full-time and part-time combined) was identified using the work history calendar. Similar to the study calendar, the work calendar recorded whether individuals were working full-time, part-time or not at all for every year since they turned 15 up to and including the current year at the time of the survey.

To examine the patterns of early life course events we use Kaplan-Meier survival curves in order to plot the change in timing across the cohorts. To model timing of leaving home we use logistic regression, using discrete-time event history. This method is appropriate given that time is recorded in discrete intervals (yearly ages) rather than in continuous format (Singer and Willet 2003). The existence of ‘ties’ in the data, when two or more subjects in the sample experience the event at the same time, would have introduced bias in parameter estimates for alternative approaches developed for continuous time data, such as Cox regression (Yamaguchi 1991). To fit the logistic regression of leaving home, the data was organized into a person-year format.

For the first model examining leaving home, a logistic model was fitted where the dependent variable was whether or not the respondent left home at time t, given no previous event occurrence. The model also included both time varying and fixed independent variables. To examine the determinants of the two different reasons for leaving home, to enter a relationship, or for other reasons, a competing risk framework was used. Two logistic models were fitted, one for each reason. In each case the dependent variable in the logistic regression was whether the person left home for that reason. As before, people who had not yet left home at the last period of observation, or age 35, were treated as censored. Individuals were also treated as censored if they left home for the alternative, competing reason.

For each model, analyses were conducted separately for males and females to account for the possible interaction effects between gender and the other variables, such as age, on the outcomes of interest.

Generational Changes in the Timing of Early Life Course Events

As we move from the period of modernization to the period of globalization and a more individualized life course, we would expect to see changes in the timing of early life course events. The changes we would expect to see are a delay in the age of leaving school as formal qualifications become increasingly important in a globalized economy, an associated delay in the timing of entry into the labour market, and a delay in the timing of leaving home as more young people stay in education longer. The following section describes the pattern in timing of these events using Kaplan-Meier survival curves.

For exit from full-time education we see a dramatic pattern of delay across cohorts and a convergence in the pattern between men and women (Fig. 4.1). There is a marked difference in the age of exiting full-time education between men and women born before 1955. On average there is a two year difference with women leaving before men.

Fig. 4.1
figure 1

Timing of exit from full-time education, by birth cohort and sex. (Note: Figures show the proportion remaining in full-time education at each age)

Full size image

The 1955–1964 cohort stayed in school longer with women showing the same pattern of school leaving as men. At age 16, 50% of the cohort had left school. The 1965–1974 cohort are very similar to the 1955–1964 cohort with slightly greater school retention at ages 15 and 16, particularly for women. The 1975–1984 cohort experienced a marked delay in school leaving before age 18 compared with previous cohorts. In this cohort 60% were still in school at age 17, compared with 40% of the previous cohort. There is also greater retention in the tertiary education years (ages 18–21). For men in this cohort there is evidence of an even longer prolongation of education through the early 1920s. The most recent cohort, those born from 1985, continues the trend of delaying leaving education with a majority still in full-time education at age18.

The pattern of entry into the labour market shows that, apart from some modest delays at ages 16 and 17, there is very little change in the timing of entry into the labour market. However, if we split labour market entry into first full-time and first part-time entry we see a profound change in the labour market behaviour of young people over the past half century. Across the cohorts there has been a delay in the pattern of entry into full-time employment corresponding to the delay in exit from full-time education (Fig. 4.2). The two most recent cohorts display the most dramatic changes in timing of entry into the full-time labour market. Women in the pre-1955 cohort entered the full-time labour market earlier than men. As the participation of women in education at secondary and post-secondary levels increases, we see this trend changing with men entering the full-time labor market earlier. For the most recent cohort, there is no discernible difference between the pattern of entry to full-time employment between men and women.

Fig. 4.2
figure 2

Timing of entry into the full-time labour market, by birth cohort and sex. (Note: Figures show the proportion entering full-time employment at each age)

Full size image

Part-time employment has become a more important labour market activity for young people across the cohorts. For the pre-1955 cohort only 50% experienced part-time employment by age 25. For the 1975–1984 cohort 88% had worked part-time by age 25 and 88% of the 1985+ cohort had worked part-time by age 20.

These changes in education and labour market activity clearly show a pattern of delayed exit from full-time education and delayed entry to full-time employment. There has also been convergence in the experience of men and women so that there are virtually no differences between men and women in education and employment experiences in the late teens and early twenties. Another change over this time period is the increasing importance of part-time employment. For the majority of part time workers this work is occurring alongside full-time education leading to increasing complexity in this early part of the life course (Fussell et al. 2007).

The survival curves for leaving home show a continual pattern of delayed home leaving (Fig. 4.3). There is little change for men between the first two cohorts but evidence of delay is seen through the changing pattern of home leaving for women particularly after age 20.

Fig. 4.3
figure 3

Timing of leaving the parental home, by birth cohort and sex. (Note: Figures show the proportion entering full-time employment at each age)

Full size image

By the 1975–1984 cohort some divergence between men and women reappears as men delay leaving home into their mid-twenties. For women the pattern of leaving home in their early twenties, is more rapid than men, but has still been delayed across the cohorts. The 1985+ cohort shows a significant delay in home leaving compared to the previous cohorts and the pattern for men and women is nearly identical up to age 20. There is evidence of the beginning of a diverging pattern at age 20 that would correspond to the differences between men and women seen in the previous cohort. However, it remains to be seen whether this cohort follows previous cohorts or if they forge a new pattern of timing of home leaving.

Generational Changes in the Reason for Leaving Home

The delayed timing of leaving home across the cohorts (Fig. 4.3) corresponds closely to the increased time spent in education as shown in Fig. 4.1. However, the different patterns for women are not fully explained by differences in education as the educational patterns for men and women have converged. The other reason suggested for changes in home leaving is the changing timing and nature of relationship formation. This section investigates the changes in the reasons for leaving home. The reasons for leaving home are classified as either to enter a live-in relationship, or for any other reason.

Across the cohorts, there is a decline in the proportion of people who leave home to directly enter a relationship (Fig. 4.4). For men this decline occurs between the first two birth cohorts and then plateaus. Close to a fifth of the pre-1955 cohort left home to enter directly into a live-in relationship. This fell to 9% for the 1975–1984 cohort. For women there has been a more gradual decline starting with close to a quarter of women in the pre-1955 cohort leaving home to enter a live-in relationship to 15% of the 1975–1984 birth cohort.

Fig. 4.4
figure 4

Leaving the parental home to enter a relationship, by birth cohort and sex

Full size image

While we can identify a change in the propensity to leave home to enter a live-in relationship, the nature of the relationships entered after leaving home has seen an even more dramatic change. Table 4.1 shows the distribution of type of relationship, whether a cohabiting union or marriage, for those who moved directly from the parental home to a relationship. There is an almost complete reversal in the type of first live-in relationship where people moved directly from the parental home into the relationship. For the pre-1955 birth cohort, 11% of the people who left home to enter a live-in relationship entered into a cohabiting union. This figure increased steadily across each cohort to 73% in the 1975–1984 cohort.

Table 4.1 Relationship type of directly entered union (%), by birth cohort and sex
Full size table

Over time young people have become a bit more likely to leave home for reasons other than relationship formation. However, the most dramatic change is the type of union they move to. These findings go against the idea that young people have been increasingly likely to live independently before forming co-residential relationships. They are moving into live-in partnerships at a slightly lower rate, and at a slightly higher age, but the relationships are now much more likely to be cohabitations rather than marriage. This may reflect a desire for less permanent and flexible relationships in the early adult years, saving marriage till later in the life course.

Factors Associated with Changes in Leaving the Parental Home

The final section of this chapter examines the factors associated with leaving the parental home. In order to look at factors associated with leaving home we run three discrete-time event history logistic regression models. All three models control for a variety of factors found to be associated with leaving home. Birth cohort and age left home are used to control for the changing impact of social and economic change on the timing of leaving home. Respondents education and parental education and occupation are used to account for differences in socio-economic status, while parental birthplace (non-English speaking country or not) controls for differences in cultural background as well as other factors that may be associated with migration and settlement. City of residence when growing up (in the teenage years) was included to examine differences that may be due to rural/ urban differentials. These could be either attitudinal or practical. For instance it may be necessary to leave home early in rural areas to continue education or to seek employment. We also include whether or not the respondent was working or studying in each year.

The first model (Table 4.2 columns 1 and 2) compares leaving home to staying at home. We see fairly similar patterns for both males and females. As expected, each successive birth cohort left home later. The odds of leaving home for individuals in cohorts born before 1965 are greater than the odds of leaving home for those born between 1865 and 1974. Similarly, those born in cohorts after 1974 show a decreasing likelihood of leaving home. Age left home shows that as age increases the likelihood of leaving home also increases. However, there is a decrease for women aged 25 years or over.

Table 4.2 Factors associated with leaving home (odds ratios)
Full size table

Socio-economic status has an impact but is not as clear as might be expected and significance levels vary. Higher levels of education are associated with increased odds of leaving home and post secondary education of mothers is associated with increased propensity to leave home for females. Having a parent born in a non-English speaking country is associated with decreased likelihood of leaving home.

Location when growing up shows a fairly linear association with leaving home. Compared to growing up in a major capital city, the likelihood of leaving home rises as the location becomes smaller and more remote. Working is associated with increased likelihood of leaving home and studying full-time is associated with a decreased likelihood of leaving home.

The second and third models are competing risk models that compare leaving home to enter a relationship with leaving home for other reasons. We find the reason for leaving home is associated with different factors. First, age at leaving home is associated with leaving home to partner but not for leaving home for other reasons. Second, completed level of education is associated with leaving home to partner for women, but not for men. The odds of leaving home to go directly into a live-in relationship rise for women as level of education rises. While the education level of men is not associated with leaving home to partner it is associated with leaving home for other reasons. Men’s likelihood of leaving home increases as their education level increases.

Location when growing up appears to be associated with leaving home for other reasons, and not to partner. This would lead to speculation that it is more likely to be opportunity driven than to be associated with different attitudinal factors.

And finally, employment increases a woman’s likelihood of leaving home to partner and a man’s likelihood of leaving home for other reasons. Studying full-time is associated with later home leaving for other reasons for women but not for men.

Discussion

This chapter explores the earliest stages of the adult life course and charts changing patterns in the transition to adulthood of young Australians. The chapter uses information from the Negotiating the Life Course survey both retrospectively and prospectively to explore the timing of school, work, and leaving home transitions. It is apparent that successive cohorts of young Australians are experiencing similar changes in the life course that have been experienced in other Western industrialized societies. First, time in education is being prolonged and men and women are now sharing the same educational timing experience. Second, entry into the full-time labour market has been delayed, corresponding to the increase in education. However, part-time employment has become part of the normal experience for young people and is usually combined with full-time study. As the period of time in education is prolonged, the transition out of the parental home is also being delayed. This change is associated with reduced financial resources due to later full-time labour market entry. Across the cohorts it is becoming less common for people to leave home to enter directly into a live-in relationship. Even more dramatic though, is the form of the relationship entered when leaving home to enter a relationship. These are now most likely to be in the form of a cohabiting union, rather than a marriage. This reversal of relationship types has occurred only in the cohorts born in the second half of the twentieth century.

The competing-risk event history models used in this paper show that the factors that influence home leaving to enter a relationship are different to those where home leaving is for another reason. These finding show similar patterns to those found in the US, Netherlands and West Germany (Mulder et al. 2002). Education is associated with delayed home leaving for women. However, when a woman does leave home she is more likely to leave home to partner than to live independently. Education is also associated with delayed home leaving for men, but men are leaving home for reasons other than to partner irrespective of their level of education.

The other strong relationship found was for residential location during the high school years. Location when growing up is associated with leaving home for other reasons but not to partner. This is true for both men and women. This finding suggests that people are more likely to leave home for reasons such as study and to find employment if they do not live in a major urban area. While educational opportunities have increased over the past two decades, in regional areas of Australia there is still an overwhelming probability that a young person will need to leave home to attend a tertiary institution if they live in small or remote locations. Once finished education, there is also a strong pull away from smaller locations in order to enter the labour market, particularly for those with tertiary qualifications.

The findings suggest that the changes in behavior identified in the literature on the second demographic transition and globalization can both be seen to be impacting on the structure of the early life course. For example, ideational change has made it possible for young people to choose cohabitation over marriage when leaving home to enter a live-in relationship. However, institutional constraints have made it more important for young people to prolong education, to work part-time while studying, and to leave home to study when higher educational opportunities are not available, such as outside major urban areas, and to delay family formation due to the increasing demands of study and work. But, while the institutional structures imposed by a globalizing society may impact on an individual’s choices, their range of choices are only available due to the ideational change that has already been occurring.