In Australia, as in most Western societies, there have been dramatic changes in family formation patterns in recent decades. Australia has witnessed a rise in cohabitation prior to marriage; a delay in the age at first marriage and an acceptance of remaining single as a legitimate choice; a decline in fertility and an increase in the age at first birth; a rise in ex-nuptial births (particularly amongst older women); and an increase in divorce (which stabilised in the 1980s, but increased slightly in the 1990s).

In comparing recent demographic trends on some major family formation measures in Australia and six other Western countries, McNicoll (1990) states that what appears to be occurring in Western countries is a 'convergence to diversity'.

Demographers now refer to the 'second demographic transition' (Lestaeghe 1991), with these recent changes being seen as part of long-term historical trends linked to modernisation. There has been a rise in secularism and an increasing emphasis on individual aspirations. The increased participation of women in the labour force and their virtual achievement of gender equality has made them less economically dependent on men, and economic theorists link demographic changes in relation to marital timing with the changing opportunity costs of women (Oppenheimer 1988; Becker 1981). Technological advances, too, have led to change, with improved contraception and inexpensive and simple sterilisation procedures having a considerable impact on fertility control.

Accompanying these changes have come shifts in values about families. Australian Institute of Family Studies research found there is a wider tolerance of shifts away from traditional family forms and more egalitarian attitudes towards sex roles (Glezer 1984).

AIFS AUSTRALIAN FAMILY FORMATION STUDY

The Institute's longitudinal Australian Family Formation Study, which surveyed a national sample of young Australian aged 18-34 years in 1981 and re- interviewed 1500 of them ten years later in 1991, provided an opportunity to look at how, why and when Australians form or do not form families.

The study enabled us to look at first marriage trends. At the time of the second interview in 1991, the ages of respondents were between 27-43 years - an appropriate age range for examining the timing of marriage and cohabitation prior to first marriage. According to the Australian Bureau of Statistics, the median age at first marriage in 1991 for men was 26.7 years and for women 24.5 years, so respondents who had not yet married at the time of the 1991 interview could be said to be delaying marriage.

Each generation experiences different historical conditions while growing up, thus creating different socialisation experiences for each age group (Ryder 1965). Growing up in times of social upheaval and unparalleled opportunity, the post-second world war generation being studied here has been at the forefront of the changes in gender roles, and family formation changes. However, of this group of 'baby boomers', those born immediately after the war had different socialisation experiences from those born in the late 1950s and early 1960s. For this reason, this paper compares respondents who, at the time of the 1991 interview, were aged 35 years and over with those aged under 35 years.

PATHWAYS TO FAMILY FORMATION

Until recently, it was expected that prior to marriage young people lived at home with parents. The ideal was to marry and then have children, although in the 1960s around one in four brides were pregnant at the time of marriage. Of course, there have always been those who deviated from this path, historically often in quite large numbers, but this has not been the case until recently this century. Today young people are leaving home to live independently from their family of origin before settling down and marrying. Increasingly, they are cohabiting informally in marriage-like relationships prior to marriage and having children outside of marriage. In other words, the pathways to forming families and the sequence of family formation events is more unpredictable than in the past.

LEAVING HOME

In 1991, only 1.5 per cent of respondents reported they were still living with their parents and had never left home. The median age of leaving home for men was 20 years and 19 years for women. There were no significant differences in the median age of first leaving home between the older and younger age groups.

Reasons given for leaving home the first time indicate a push for greater independence among the younger generation, with those under 35 years being less likely to leave for education, vocational reasons or for marriage than those over 35 years. Many of the younger group returned to live with parents after initially leaving for a period of independence - a continuation of the 'revolving door' pattern that Young (1987) found in her study of young people's exit from their family of origin.

THE EVER MARRIED

In 1991, 64 per cent of men and 67 per cent of women were in their first marriage, and 14 per cent of men and 19 per cent of women had been previously married (that is, separated, divorced or remarried). The remaining 23 per cent of men and 13 per cent of women had never married.

As Figure 1 shows, of those who had ever married, 64 per cent had followed the traditional path of marrying without living together or having an ex- nuptial birth, 29 per cent had cohabited prior to marriage, and 7 per cent had cohabited and/or had an ex-nuptial child.

Figure 2 shows that when those aged 35 years and over are compared with those under 35 years, 71 per cent of the older group compared with 56 per cent of the younger group followed the traditional path. There has been a substantial rise in the numbers cohabiting prior to marriage in the group aged under 35 years, but very little difference in the numbers having children prior to marriage.

THE NEVER MARRIED

In 1991, 23 per cent of men and 13 per cent of women had never married. However, of these, over half had lived in a de facto relationship or cohabited at some stage; 24 and 21 per cent of never married men and women respectively were cohabiting at the time of interview.

Figure 3 shows that of those who had never married (17 per cent of the total sample), 49 per cent had never cohabited, 38 per cent had cohabited, and 13 per cent had cohabited and/or had a child (very few had children outside a relationship).

Ex-nuptial birth rates have been steadily climbing in Australia since the 1960s: in 1991, 22 per cent of births were ex-nuptial compared with 9 per cent in 1971. However, according to the Australian Bureau of Statistics, in about eight out of ten cases paternity is acknowledged, which suggests that, in the majority of cases, these are children born into cohabiting relationships.

In the first interviews in 1981, only 10 per cent of never married respondents expected not to marry. In 1991, of the 17 per cent who had never married, one in five expected not to marry. Thirty-nine per cent of men and 30 per cent of women expected that they would marry. Half the women and 31 per cent of men were uncertain as to whether they would marry, saying such a decision depended on meeting the right person. This suggests that respondents were not so much rejecting marriage as delaying the event because they had not found suitable partners. In 1991, about half the never married respondents said they would consider having a child outside of marriage.

THE TIMING OF MARRIAGE

In Australia between the 1940s and 1970, as in other Western countries, early and near universal marriage was the norm. More than 90 per cent of Australian women of marriageable age in that period married, often at a very young age. After reaching its lowest point in the mid-1970s, the age at first marriage and the proportion marrying started moving towards the levels that prevailed prior to the 1940s (McDonald 1990).

Figure 4a shows the proportions of men and women aged under 35 years and those 35 years and over who married by their 20th, 25th and 30th birthdays. As expected, there are gender differences in the proportion of respondents married at each age. Very few men in either age group were younger than 20 years when they married; the proportion of women who were younger than 20 years when they married differs little between the two age groups.

However, among both age groups for both sexes, there are significant differences in the proportion of respondents who married by age 25 and by age 30 years, clearly illustrating that the more recent generation were deferring marriage.

Bumpass and Sweet (1989) raised the issue that 'the beginning of cohabitation may be a more apt marker of union formation than the marriage date'. As a high proportion of ever married respondents had cohabited with their spouse prior to first marriage, it was decided to substitute age at marriage with the age respondents were when they first lived with their spouse before marriage (Figure 4b) and compare the results with those in Figure 4a.

When this is done the significant differences in the proportion married by different ages among the older and younger groups disappears. It appears that while people delay legally tying the knot, those in the younger group started living together at around the same age as they would have married in the past.

To a large extent the key to understanding the changes in the timing of marriage lies in a better understanding of premarital cohabitation.

COHABITATION

In Western societies, cohabitation has been a growing phenomenon for more than two decades. However, it should be remembered that, historically, 'common law' marriages were the norm, formal marriage being the preserve of the propertied classes. De facto relationships are thus of interest only as a departure from what has become 'normal' as first the Church and then the State began to control the social mechanism we call marriage.

As we have seen, since the 197Os the trend has been for young people to leave home and live independently for a period before marrying. In many instances, this involves being sexually active and setting up house in marriage- like relationships. Marriage is therefore unlikely to be a young person's first experience of living with someone in a committed sexual relationship, and, increasingly, young people are postponing marriage in favour of living together until their mid to late twenties.

There are various interpretations of cohabitation in the literature. It is regarded as a stage in courtship, a trial marriage, 'a prelude to marriage, not a substitute', as Carmichael (1990) expresses it, or as an alternative to marriage (Bumpass and Sweet 1989; Khoo 1986; Sarantakos 1984). Some view cohabitation as part of the courtship process and therefore more related to premarital dating relationships than to marriage (Rindfuss and VandenHeuvel 1990).

Results from the Institute study suggest that cohabiting relationships are as distinct from non-cohabiting relationships as they are from marriage. A gradation of relationships - that is, a range of choices - is consistent with the general trend towards diversity and individual autonomy (Glezer et al 1992).

There is increasing research interest in understanding premarital cohabitation, particularly as a growing body of research in Western countries indicates a strong association between premarital cohabitation and the breakdown of first marriages. Axinn and Thornton (1992) suggest that the experience of cohabiting significantly increases young people's acceptance of divorce.

The Institute study found that premarital factors contributing to the breakdown of first marriages included having an ex-nuptial child, premarital cohabitation, and leaving home at an early age (Glezer et al 1992).

FACTORS INFLUENCING COHABITATION PRIOR TO MARRIAGE

There has been considerable research into whether or not people's family backgrounds influence their later family formation behaviour, such as premarital cohabitation. Thornton (1991) found that parental separation and divorce increases the likelihood that children will enter non-marital cohabitation relationships, and Thornton and Camburn (1987) found that maternal attitudes and behaviour can influence the premarital sexual behaviour of daughters. It has also been found that those most likely to cohabit tend to have higher education levels, come from non-religious families, and have lived independently from parents (Khoo 1986).

There is still much to learn about the antecedents of cohabitation. The Institute study provided an opportunity to examine some of the family background and early adult experiences that influence or are associated with premarital cohabitation. Regression analyses revealed that people who cohabit tended to be the younger respondents, particularly those from urban rather than rural areas, and were not religious. They also tended to have left home early (under 18 years) and had sexual relations prior to 18 years. For women, tertiary education was associated with cohabitation.

As it is also likely that other aspects of family background or perceptions of family dynamics contribute to family values and influence future family formation behaviour, the study included measures of these dimensions. It was found that women who cohabited were likely to have grown up in families where parents were more permissive and placed no restrictions on young adults leaving home or engaging in premarital sexual relations. Conversely, women from families which discouraged leaving home and prohibited premarital sexual behaviour were unlikely to have cohabited. For men, perceived parental values did not have a significant effect, no doubt because parents are likely to be more concerned over their daughters' sexual behaviour than their sons' - as Rubin (1991) has noted, sexual double standards still exist!

Both men and women who were unhappy living with their family, who did not feel accepted by their parents and felt there was a high level of tension in the home were far more likely to have left home early and to have cohabited than respondents who reported a happy family upbringing. It was found that men, but not women, who experienced the breakdown of their parents' marriage were more likely to cohabit.

These results indicate that there is a strong link between family background experiences and young adult cohabitation.

CHARACTERISTICS OF THOSE WHO HAVE COHABITED

Respondents who had cohabited compared with those who had not cohabited prior to first marriage tended to have liberal family values, and were also likely to value marriage and children less than those who had not cohabited. They had egalitarian beliefs about sex roles, which is also reflected in how they shared household tasks.

Increasing individualism and the desire for personal autonomy are reasons often given as part of the explanation of the major socio-demographic changes in recent decades (Lesthaeghe1991). As McDonald (1988) states: 'The trend for young couples to live together rather than marry can be seen as an experiment in the pursuit of personal autonomy. Young people want companionship, a sexual life and autonomy from parents. These could be provided through marriage, but marriage is still seen as a lifetime commitment and hence a massive incursion on personal autonomy.'

The study included a scale that measured individualism in relationships: it looked at independence in relationships, and the issue of personal growth versus commitment to marriage. Those who had cohabited were significantly more likely to score highly on this individualism measure than those who did not, with women scoring more highly than men.

While those living in cohabiting relationships at any given time are a small proportion of the population, just under half those under 40 are likely to cohabit at some stage in their lives.

There is great diversity in the kind of cohabiting relationships couples form, and motivations vary between couples and within couples, and, no doubt, change over time. Reasons given for cohabiting are emotional, economic and pragmatic, and levels of commitment vary, as does the intention to formalise the relationship.

Cohabiting relationships are not only distinct from both marriages and non- cohabiting relationships, they also vary considerably within themselves. This indicates clearly the extent to which the conflicting ideals of intimacy and autonomy have produced relationships individually designed by the couple to suit their own particular needs.

SINGLES LIFESTYLE

'Never married' is a convenient demographic classification, but those who are never married are a very diverse group in terms of how they live their lives. Some will never marry, a small number reject marriage in favour of cohabitation, some will have a relationship where they live together part of the time, others who are homosexual will live in relationships and be denied access to legal marriage, others will live alone by choice or because they have not found the right mate.

In the Institute study, among the never married in 1991, 57 per cent of men and 55 per cent of women were not in a relationship, 19 per cent of men and 24 per women stated they had a relationship but were not living with their partner (although many 'slept over' some of the time), and 24 per cent of men and 21 per cent of women were cohabiting.

Since the mid-1970s there has been increasing acceptance of people who choose not to marry. How do these singles live their lives?

Around four in ten never married single respondents said that meeting potential partners was a problem, with about half of those without a partner saying they would like to have one; 23 per cent of men and 31 per cent of women said they were not interested in forming a relationship; and the others hoped that a current friendship would develop into a deeper relationship.

The main way singles meet potential partners is through friends (men 76 per cent, women 83 per cent) and through work (men 42 per cent, women 52 per cent). Just over a third find holidays, and going to clubs, pubs and discos provide opportunities for meeting potential mates. Men seek partners in their outdoor activities (35 per cent) and when they attend sporting events (22 per cent); women are significantly less likely than men to mention outdoor activities (17 per cent ) or sporting events (7 per cent).

One of the main reasons why never married people are not in relationships, for men particularly, is that they find it hard to get close to people - 53 per cent of men compared with 35 per cent of women find intimacy a problem. Women (53 per cent) are more likely than men (43 per cent) to say they have been hurt in the past and feel very wary when it comes to relationships.

While those people living a single lifestyle are often happy with life as it is (women 67 per cent and men 52 per cent), eight out of ten respondents said they had just not met the right person. Singles tend to have a lot of investment in their jobs and work life (55 per cent of men 46 per cent of women) which means they put less emphasis on their personal relationships. A few are getting over past relationships, and are not yet ready to risk themselves in the relationships stakes.

The reality of the 1990s, with the AIDS virus and the high incidence of sexually transmitted disease, is a factor single people consider when seeking partners in four out of ten cases (for both men and women).

Generally, never married respondents recognised that being single allowed them to be independent (85 per cent of women compared with 77 per cent men).

'I am economically independent, secure employment wise, and travel regularly. No plans to ever marry.' (male)

'I would like to, but the longer you leave it before marrying the more freedom you are used to, and you think twice about giving it up.' (male)

'I'm too independent. It would be hard to share a house and money with someone. I've become too self-centred.' (female)

'Not interested - I don't need it. I find the idea of committing for a lifetime unrealistic.' (female)

Sixty-two per cent of single men said that being single involved fewer worries and responsibilities. Only half the single women in the sample agreed with this. However, women (67 per cent) were more likely to believe that being single provides more time to experience life and find out about yourself than single men (60 per cent). More women (66 per cent) compared with men (57 Per cent) also recognised that in career terms being single enabled them to spend more time at work to get ahead. This is understandable as marriage usually involves children who impinge far more on a woman's career than on a man's. Women who delay marriage are more likely to have had higher education, and as a result tend to have high expectations about their career opportunities.

What about the downside of single life? Forty-one per cent of women and 54 per cent of men said that it was lonely being single, and 42 per cent of women and half the men believed you miss not having children when single. Just under six in ten single men and women believed that close friendships do not make up for not having a partner.

However in general, single life is seen as positive by single men and women. The women (76 per cent) saw far more advantages to being single than the men (56 per cent), and seemed to lead more satisfying lives. Perhaps this is because women find it easier than men to form intimate friendships and relationships.

Women were more likely than men to express their wariness about making a commitment to marriage. One woman said such a notion made her feel 'insecure'. Another said she 'didn't expect to meet someone I would want to marry'. As one woman summed up:

'It depends on meeting a suitable compatible person, who also wishes to marry. I meet lots of men but very few with the qualities I am looking for. I believe that men have a little bit of catching up to do and need to be encouraged to be more sensitive and open to their feelings, something which has been discouraged in the past.'

While there is strong support for marriage as an institution, those who have never married are more likely not to support it than those who have married (30 per cent of never married men and 23 per cent of never married women compared with 15 per cent of married men and 10 per cent of married women).

'The older I get, the less inclined towards marriage I become. Particularly resulting from observations of the divorces and separations of friends and relatives.' (single female)

'Married friends don't have time to do anything - all their time is spent with their partner. Wives can be very bossy, and I'm an independent person.' (single male)

When it comes to the dilemma of one's own autonomy versus commitment to marriage and family, never married respondents were more 'individualistic' than those who were married. The majority of single respondents felt positive about their status, with women being generally happier with their lives than men.

CONCLUSION

The Institute's Australian Family Formation Study confirms that changing family formation trends such as premarital cohabitation and delayed marriage are more prevalent in those under 35 years than those over 35 years. These differences are likely to reflect the changes over time in family values and gender roles and the greater tolerance of diversity in patterns of family formation.

* Figures referred to in text are available from The Database Indexer, Australian Institute of Familiy Studies, 300 Queen Street, Melbourne 3000, reference FM34/16.

REFERENCES

• Axinn, W. and Thornton, A. (1992), 'The relationship between cohabitation and divorce: selectivity or causal influence?', Demography, Vol.29, No.3.
• Becker, G. (1981), A Treatise on the Family, Harvard University Press, Cambridge, Mass.
• Bumpass, L. and Sweet, J. (1989), 'National estimates of cohabitation', Demography, Vol.26, No.4, pp.615-625.
• Carmichael, G. (1990), 'Analysis of marriage and informal cohabitation', Australian and New Zealand Journal of Sociology, Vol.27, No.1, pp.53-72.
• Glezer, H. (1984), 'Antecedents and correlates of marriage and family attitudes in young Australian men and women', XXth International CFR Seminar on Social Change and Family Policies, Key papers, Part 1, Melbourne.
• Glezer, H., Edgar, D., and Prolisko, A. (1992), 'The importance of family background and early life experiences on premarital cohabitation and marital dissolution', Paper presented 'Family Formation and Dissolution: East and West Perspectives', Taipei, May 1991, ISSP and Sun Yat Sen Institute for Social Science, Academica Sinica.
• Lesthaeghe, R. (1991), 'The second demographic transition in western countries: an interpretation', IPD Working Paper, Interuniversity Programme in Demography, Belgium.
• Khoo, S-E. (1986), Living Together: Young Couples in De Facto Relationships, Working Paper No.10, Australian Institute of Family Studies, Melbourne.
• McDonald, P. (1988), 'Families in the future: The pursuit of personal autonomy', Family Matters , No.22, pp.40-47.
• McDonald, P. (1990), 'The 1980s: social and economic change affecting families', Family Matters , No.26, pp.13- 18.
• McNicoll, G. (1990), 'Are Australian families like others?', Research School of Social Sciences Seminar Series No.17, Australian National University, Canberra.
• Oppenheimer, V.K. (1988), 'A theory of marriage timing', American Journal of Sociology, Vol.94, pp.563-591.
• Rindfuss R. and VandenHeuvel, A. (1990), 'Cohabitation: a precursor tomarriage or an alternative to being single?', Population and Development Review, Vol.16, No.4, pp.703- 726.
• Ryder, N.B. (1965), 'The cohort as a concept in the study of social change', American Sociological Review, Vol.30, pp.843-861.
• Sarantakos, S. (1984), Living Together in Australia, Longman Cheshire, Melbourne.
• Thornton, A. (1991), 'Influence of the marital history of parents on the marital and cohabitational experiences of children', American Journal of 'Sociology, Vol.96. No.4, pp.868-894.
• Thornton, A. and Camburn, D. (1987), 'The influence of the family on premarital sexual attitudes and behaviour', Demography, Vol.24, No.3, pp.323-340.
• Young, C. (1987), Young People Leaving Home in Australia, Monograph No.9, Research school of Social Sciences, Australian National University, Canberra.