Introduction

As in many industrialized countries, Australia’s population is ageing. The proportion of the population aged 65 and over is projected to increase from 14% in 2011 to 21% in 2031 (ABS 2008). Australia is also a country of immigration, with 22% of its population in 2006 being foreign-born. The percentage foreign-born is even higher in the age group 65 and over, at 31% in 2006. This is because the immigrant population has an older age structure than the Australian-born population. In 2006, 19% of the overseas-born population were aged 65 and over compared with 11% of the Australian-born population. The number and percentage of the immigrant aged are also projected to increase (Gibson et al. 2001) and they constitute an important dimension of Australia’s population ageing (Borowski and McDonald 2007).

Because of the different waves of immigration during the second half of the twentieth century, some immigrant groups have relatively high proportions in the older age groups. The 1950s and 1960s were years of high immigration from Southern and Eastern Europe: a large proportion of immigrants from these countries are now aged over 65. For example, 52% of Australian residents born in Italy were aged 65 and over in 2006; as were 45% of those born in Greece, 51% of those born in Hungary and 37% of those born in Poland (Table 1). With the end of the White Australia policy in the early 1970s and its restrictions on immigration from non-European countries, Australia began receiving immigrants from other geographic regions, particularly from Asia. Among these more recently arrived immigrants, the proportion in the older age groups is much lower, but the number of older people in some of the larger immigrant communities is still large. Examples are immigrants from China, of whom over 25,000 were aged 65 and over in 2006 (although they form just 12% of the total birthplace group), immigrants from India with nearly 15,000 aged 65 and over (10% of the total birthplace group), and immigrants from Vietnam with over 11,000 (7% of the birthplace group) (Table 1). There is consequently increasing interest in the immigrant aged and their social and economic well-being.

Table 1 Number of people and percentage aged 65 and over, by birthplace, Australia 2006
Full size table

The paper examines differences in the social and economic well-being of the immigrant aged by country of origin as measured by a number of indicators, and the factors affecting their well-being. The data analysis is guided by the concept of the ‘Third Age’ as a life-cycle stage of independent living after age 65, focusing on the absence of core activity restrictions and the pursuit of an active retirement. The indicators examined include living arrangements, proficiency in English, level of education, income, assistance with the activities of daily living, and connectedness to family and community through paid work, volunteering, and caring for family members and others.

Previous research on the immigrant aged in Australia

There was much interest in the immigrant aged population in Australia in the 1980s when there was a realization that immigrants who arrived in Australia during the postwar years would be moving into the older age groups in the coming years. Some of the Eastern and Southern European birthplace groups already had a median age of 50 years or more. In 1984, the Australian Institute of Multicultural Affairs conducted a major study of the ethnic aged based on a survey of over 1,100 persons aged 60 and over from six birthplace groups, China, Germany, Greece, Italy, Poland and the former Yugoslavia. The study examined their experiences in relation to retirement, living arrangements, community care and use of services, and found differences within and between the birthplace groups in the pattern of these experiences (AIMA 1985, 1986). This and other studies also showed that English proficiency was relatively low in some birthplace groups, even among those who had lived in Australia for many years, and this had limited their social networks and participation (AIMA 1985; Rowland 1991). The ethnic aged were also found to have lower incomes, a lower rate of home ownership, more restricted family networks and greater likelihood of feeling isolated (AIMA 1985, 1986; Rowland 1991). However, a higher proportion was married and would have the support of a spouse. The studies point to considerable diversity in the ethnic aged; they also show some similarities to the rest of the Australian population in their preference for remaining in their own home with the support of a spouse and/or family members. These characteristics of the ethnic aged have led to the suggestion that they might be derived from life chances rather than culture or lifestyle preferences (Rowland 1991, 1997).

In 2000 the Australian Government’s Department of Immigration and Multicultural Affairs commissioned another study of the immigrant aged to update the earlier studies. Focusing on the social and the financial circumstances of the overseas-born population aged 55 and over, and based partly on data from the 1996 population census, the study found an increase in their English proficiency over time, but lower work participation and higher unemployment rates among the ethnic aged compared to the Australian-born aged. Older migrants were also more likely to be dependent on government pensions (Benham et al. 2000). The study compared the immigrant aged according to four birthplace groups based broadly on level of English proficiency, which precludes comparison by ethnicity.

The ethnic aged in these studies referred to immigrants from continental Europe as there was little immigration from non-European countries until after 1970 and therefore relatively few non-European immigrants in the older age groups. Comparisons were usually made between these European immigrants of non-English- speaking background with immigrants from the United Kingdom, Australia’s largest source of immigrants, and with the Australian-born aged. Comparisons by country of origin to examine differences by cultural background usually focused on the larger groups from Germany, The Netherlands, Italy, Greece, the former Yugoslavia and Poland. Immigrants from Western European countries such as Germany and The Netherlands were generally more proficient in English than those from Southern and Eastern European countries, and had more similar social and economic characteristics to immigrants of English-speaking background and the Australian-born.

Since the immigrant aged now include many people from non-European countries, it is now possible to compare the aged from Asian countries with those from European countries and also to examine ethnic differentials within the Asian immigrant aged by country of origin. The Australian population census in 2006 also collected information on a number of topics that are particularly relevant to the study of the social and economic well-being of older people. In the analysis of these new data in relation to the currently more ethnically diverse aged population, the present study both updates and expands on the previous studies of the immigrant aged in Australia.

Conceptual framework and research aims

Discussion of population ageing is usually focused on the dependency of the aged and the costs of meeting this dependency. In the case of the ethnic or immigrant aged, it has been suggested that they face the ‘double jeopardy’ of being aged and ethnic and the consequences of their ageing experiences can be doubly negative (Dowd and Bengston 1978 in Rowland 1991). On the other hand, it has also been argued that the cultural dimensions of ethnicity can facilitate adjustment to the psychic and physical constraints of ageing and that a better understanding of this process can be gained, for example, by examining the influence of country of origin, historical point of entry and the processes of acculturation and social integration on the experiences of the immigrant aged (Holzberg 1982). It is on this latter approach that the paper will focus in its examination of ethnicity and the well-being of the immigrant aged.

The examination of social and economic well-being of the immigrant aged in this paper is guided by the concept of the ‘Third Age’ as a new life cycle stage of productive ageing (see Laslett 1989). This concept, originating in Europe, refers to a new stage of life of personal fulfilment after retirement, before the Fourth Age of ‘true dependency and decrepitude’ sets in (Laslett 1989; Rowland 2003). Laslett (1989: 78–91) had suggested several preconditions for the emergence of the Third Age as a significant life cycle stage of ageing in modern societies. These include at least 10% of the population to be aged 65 and over; an average life expectancy showing that the majority of the population will survive beyond retirement age; a society with sufficient national wealth to support its older citizens at a comfortable standard of living; supportive attitudes to the participation of the aged in society; and cultural and educational resources to facilitate that participation. These conditions are present in Australia.

The paper uses Rowland’s (2003) redefinition of Laslett’s concept of the Third Age as a life cycle stage of independent living after age 65. This means experiencing a level of social, physical and economic well-being that enables an older person to live at home with little or no assistance with the activities of daily living. Additionally, there may be pursuit of an active retirement, as envisaged by Laslett: this includes involvement in productive activities such as paid work and volunteering. Volunteering is considered a key productive ageing activity and is related to well-being in older persons (Warburton et al. 2007). Based on the two dimensions of the ‘Third Age’, the paper examines a number of indicators of social and economic well-being among the immigrant aged to investigate the extent of ethnic diversity in these measures by country or region of origin.

It has been suggested that recency of arrival, proficiency in the language of the country of settlement and educational background also affect the ageing experiences of migrants, with ethnicity not necessarily the most important factor (Rowland 1991). Recent arrivals and immigrants experiencing difficulties with the language of the country of settlement may be less integrated into the community and more dependent on their relatives for social and economic support (Rowland 2007). Benham et al. (2000) found that coresidence with relatives was more likely among the ethnic aged where English language proficiency is relatively low. Because migrants from different countries differ in their period of arrival in Australia, English language proficiency and level of education, it is necessary to control for these variables in examining the relation between birthplace as a measure of ethnicity, and the well-being of the immigrant aged.

Data and method

The paper is based on data from the 2006 Australian population census and focuses on eight outcome measures of social and economic well-being reflecting independent living and the pursuit of an active retirement among the immigrant aged. The measures are marital status, living at home (not in an aged care institution), no need for assistance with core daily activities, income, participation in paid work, volunteering, caring for children under age 15 and caring for an older, sick or disabled person or family member.

Marital statusFootnote 1 can be an important indicator of family resources (Rowland 2003) and social well-being, particularly for older people. Older people who are married have the family support and companionship of a spouse in contrast to those who are widowed, never married, divorced or separated. On the other hand, some married older persons are also the primary carers of their spouse. Marital status is therefore also an explanatory variable in the data analysis of the other measures of well-being.

Living at home and not in institutional care is an indicator of independent living among the aged. The census has information on whether each person is enumerated in a private dwelling or a non-private dwelling. Older persons who are enumerated in a private dwelling are considered to be living at home while those who are enumerated in a non-private dwelling are assumed to be in an institution, which may be an aged care hostel, nursing home or hospital.

A direct indicator of physical independence among the aged is the lack of need for assistance with daily self care activities. In the 2006 census, four questions were asked of each person: whether he or she needs someone to help with daily self care, movement or communication, and the reasons for needing assistance. The Australian Bureau of Statistics then classifies people as needing assistance with core activities if they need assistance with any or all three types of activities because of a disability, long-term health condition (lasting 6 months or more) or old age.

Personal income is examined in the paper as an indicator of economic independence. Older people with income of their own generally have more control over their lives; they are less dependent on other family members and can afford to live on their own. The aged population is divided into two income groups in the data analysis: those with a weekly income of less than $250 and those with a weekly income of $250 or more. These two categories provide an approximate differentiation between those whose income does not exceed the government age pensionFootnote 2 at the time, which provides for a basic standard of living, and those with other sources of higher income.

Four variables measure the pursuit of an active retirement: being in full-time or part-time employment in the week before the census; volunteering for an organization or group at any time in the last 12 months; looking after a child or children under age 15 without pay; and caring for a family member or other person who has a disability or long-term illness or problems related to old age. The 2006 census was the first census in Australia to include questions on participation in volunteer work and caring responsibilities.

Two measures of human capital, English language proficiency and level of education, are also examined as measures of social and economic well-being and as correlates of the eight outcome variables. English is the language of commerce, instruction and almost all daily social interchange in Australia. Studies of immigrant settlement outcomes in Australia have shown that English proficiency and level of education are positively correlated with social, economic and community participation and integration (e.g. Cobb-Clark and Khoo 2006; Khoo and Temple 2008). Not being able to speak English well can inhibit the formation of social networks and increase the risk of social isolation among older immigrants. Education has been considered an important personal resource that may contribute to both quantity and quality of life in old age through its effect on health and lifetime earnings (Rowland 2003). Two categories of English proficiency are considered: speaking English well/very well, or not well/not at all. Level of education is also divided into two categories: with post-school qualifications and without post-school qualifications.

The descriptive part of the data analysis uses the full census data file to examine the measures of social and economic well-being for the total resident overseas-born population aged 65 and over by sex and country of birth. The focus is on the immigrant aged from non-English-speaking countries of origin with at least 5,000 people aged 65 and over. Comparisons are also made between all immigrants from non-English-speaking countries and immigrants from the main English-speaking countries, United Kingdom, Ireland, New Zealand, Canada, United States of America and South Africa, and also the Australian-born population aged 65 and over. The data analysis differentiates between two age groups: those aged 65–79 and those aged 80 and over.

Since the different birthplace groups can and do differ from one another in their sex and age distribution, English language proficiency and level of education in the older age groups as well as in the outcome measures of well-being, multivariate logistic regression is used to control for these characteristics to see whether the country of birth differences in the descriptive analysis persist. The statistical analysis uses the 1% sample unit record file from the 2006 census. The country-of-birth categories in the analysis of data from the 1% sample file are limited to six countries that have been sources of significant numbers of immigrants to Australia, with the remaining non-English-speaking countries grouped into nine regions. The multivariate statistical analysis excludes individuals whose birthplace was not stated and two relatively small regional groups, the Americas and sub-Saharan Africa, because it is not possible to separate the small number of immigrants from the non-English-speaking countries in these regions from those from the English-speaking countries. This results in a sample size of 23,227 individuals aged 65 and over (of whom 8,010 were foreign-born) for the logistic regression analysis.

Two sets of regression models were run for each well-being indicator. The first set of regressions includes the Australian-born aged as the reference birthplace category so the immigrant aged can be compared with the Australian-born aged. The second set of regressions is restricted to the overseas-born aged only and focuses on ethnic disparities within the group. The reference group here is the immigrant aged from the United Kingdom and New Zealand, the two largest sources of immigrants of English-speaking background. The second set of regressions includes year of arrival as a control variable; the coding of this variable in the 1% sample file groups all arrivals before 1995 into a single category. Therefore immigrants are grouped into only two categories in the data analysis: those who arrived before 1996, and during 1996-2006. This grouping has some relevance in examining economic well-being as immigrants are eligible for the age pension only after 10 years of residence in Australia. Thus, the immigrant aged who arrived during 1996–2006 would not have access to the age pension.

Country of birth is a good indicator of the ethnic origin of most of the immigrant aged population in Australia. Data on ancestry are also collected in the census and show a high correlation between ancestry and country of birth responses for the immigrant aged (for example, 97% of those born in Italy stating Italian ancestry), with the exception of a few countries. The 1% sample file shows that 42% of the immigrant aged born in India claimed English or Irish ancestry. However, ancestry data also have their limitations. The one that most affects this study is that the immigrant aged claiming Chinese ancestry come from many countries and have very different migration histories and social and economic characteristics depending on their country of origin. Country of birth is a more meaningful descriptor when examining the settlement outcomes of these immigrants. Parallel analyses using the ancestry variable in the 1% sample file (which has the same broad regional categories as for the birthplace variable, plus some larger ancestry groups) instead of the birthplace variable produce generally similar results.

Social and economic well-being of the immigrant aged

Table 2 shows the indicators of social and economic well-being reflecting independent living and the pursuit of an active retirement for the immigrant aged from non-English-speaking countries compared with those from the main English-speaking countries and with the Australian-born aged. Table 3 highlights some of the differences by country of birth by showing the indicators for immigrants from selected European, Middle Eastern and Asian countries.Footnote 3

Table 2 Measures of social and economic well-being of the immigrant aged compared with Australian born aged
Full size table
Table 3 Social and economic well-being of the immigrant aged: selected countries of origin
Full size table

Compared to the Australian-born aged, the immigrant aged from non-English-speaking countries have a higher proportion currently marriedFootnote 4 and a lower proportion in institutional care. They also have lower proportions with post-school qualifications, in the higher income group and not needing assistance with daily activities than the Australian-born aged. The measures of the pursuit of an active retirement also show a lower proportion of them in paid work or volunteering compared with the Australian-born aged, although they have a higher proportion caring for children. There is not much difference between the immigrant aged and the Australian-born agedin the proportion caring for others.

Table 3 shows some consistent patterns of difference in many of the measures of social and economic well-being by country of origin of the immigrant aged. One pattern of difference is between different European birthplace groups, with immigrants from Western European, Eastern European and Southern European countries showing differences in education, absence of restrictions with core activities, income, home ownership, participation in paid work, volunteering and care of children. The proportion not needing assistance with daily activities is higher for immigrants from Germany and The Netherlands than for immigrants from the countries of Southern Europe such as Italy and Greece. English language proficiency is much lower among the immigrant aged from Italy and Greece than those from Western European countries such as Germany and The Netherlands. The immigrant aged from Italy and Greece also have much higher proportions married and lower proportions divorced or separated (data not shown but 4–6% in the 65–79 age group) than those from Germany and The Netherlands (11–12%).

Asian birthplace groups also differ from one another. The immigrant aged from Malaysia and the Philippines had higher proportions with post-school education, higher income, participation in paid work and volunteering than those from Vietnam and China. The immigrant aged from Vietnam, China and the Philippines also have a lower proportion not needing assistance with daily activities and a lower rate of home ownership than those from Malaysia. Older immigrants from China and Vietnam also have relatively low levels of English proficiency. These migrants had come from countries where English is not a familiar language. Their low level of English language proficiency suggests that their social networks would be restricted to their families and ethnic community. In contrast, higher proportions of the immigrant aged from Malaysia and the Philippines are proficient in English, having come from countries where English is widely spoken. The immigrant aged from Vietnam appeared to be the most disadvantaged on a number of measures of social and economic well-being.

The lower rate of participation in paid work in some of the birthplace groups may be the result of early retirement from low-skilled work and low English proficiency. The proportion involved with volunteer work was low in birthplace groups where English proficiency was also low, which was as expected. Immigrants from The Netherlands and Malaysia, most of whom could speak English well, had higher proportions volunteering in the community. Previous studies have also found a lower rate of volunteering by the overseas-born population compared with the Australian-born population (ABS 2001; Volunteering Australia 2007). Lack of English language proficiency has been identified as the most common barrier to volunteering by migrants themselves and the organizations involved in volunteer work (Volunteering Australia 2007).

To investigate whether the observed differences by country of origin are related to differences in the immigrant aged’s demographic characteristics, level of education and English language proficiency or whether they reflect ethnic differences and other characteristics associated with country of origin, multivariate logistic regression analysis is carried out to control for the effects of these characteristics. The findings are discussed in the next section.

Multivariate logistic regression results

The logistic regression results for eight indicators of social and economic well-being relating to independent living and pursuit of an active retirement are shown in Tables 4, 5, 6, 7.

Table 4 Logistic regression results (regression coefficients and standard errors) for marital status and living at home
Full size table
Table 5 Logistic regression results (regression coefficients and standard errors) for absence of core activity restrictions and income
Full size table
Table 6 Logistic regression results (regression coefficients and standard errors) for participation in paid work and volunteering
Full size table
Table 7 Logistic regression results (regression coefficients and standard errors) for care of children and disabled/sick/old person
Full size table

Table 4 shows the results on being married and living at home (not in an institutional care facility). Differences in the probability of being married by birthplace are observed even after controlling for age and sex and human capital characteristics. The immigrant aged from Italy and Greece were more likely to be currently married than other immigrants and the Australian-born aged. Differences by birthplace in the proportion living at home and not in institutional care also remain after controlling for demographic and human capital characteristics. The immigrant aged from Southern and Eastern European and Asian countries were all more likely to be living at home and not in an institutional care facility than immigrants from the English-speaking or Western European countries and the Australian-born aged. The statistical analysis shows no difference between men and women in the propensity to live at home. However, older people who were married were more likely to live at home than those who were widowed or not married. The immigrant aged who had higher education or who were proficient in English were also more likely to live at home and not in institutional care than those with less education or whose English was not good.

Differences in the likelihood of needing assistance with core daily activities between the immigrant aged—with the exception of those from the Middle East and Pacific region—and the Australian-born aged were not significant after taking account of their demographic and human capital characteristics (Table 5). The immigrant aged from the Middle East and the Pacific region were the most likely to need assistance with daily activities, implying poorer physical health. This is consistent with the descriptive results showing the relatively low percentage of the immigrant aged from Lebanon not needing assistance (Table 3). Older migrants from the Southern and Eastern European countries were also more likely to need assistance with daily activities than the immigrant aged of English-speaking background. Many migrants who arrived in the 1950s and 1960s from Southern and Eastern Europe had worked in physically demanding jobs in construction and manufacturing and this may have affected their physical health in old age. The data analysis also shows that better physical health in old age was associated with being married, being proficient in English and having higher education.

The indicator for economic independence confirms the earlier descriptive findings that the immigrant aged from non-English-speaking countries were more likely to be in the lower income group than the immigrant aged from the main English-speaking countries and the Australian-born aged (Table 5). These findings are also consistent with those of previous studies (AIMA 1986; Rowland 1991; Benham et al. 2000) The regression coefficients also indicate a similar pattern of disparities by birthplace shown earlier in the descriptive results, that the immigrant aged from Vietnam, China, Middle East (including Lebanon) and Southern European countries were less likely than other immigrants to be in the higher income group.

The regression results for the four measures of pursuit of an active retirement are shown in Tables 6 and 7. The immigrant aged from Europe, aside from Italy and Greece, were significantly less likely to be working than the Australian-born aged even after controlling for English proficiency and education (Table 6). However, differences between immigrants from China, Vietnam and other Southeast Asian countries and the Australian-born aged or immigrants from the UK and New Zealand were not significant after adjusting for differences in demographic and human capital characteristics, suggesting that the low employment rates for the immigrant aged from Vietnam and China observed earlier in the descriptive results were likely to be related to their lack of English proficiency and, in the case of the immigrant aged from Vietnam, their lower education. The regression results show significantly higher employment among older immigrants from Northeast Asia aside from China. This group includes people from Hong Kong, Korea and Taiwan, some of whom may have been business migrants. Italian and Other Pacific immigrants also have a higher probability of working after controlling for demographic and other factors.

The lower rates of volunteering shown for the immigrant aged compared to the Australian-born in the descriptive results are confirmed in the statistical analysis. Disparities among the immigrant aged by birthplace remain even after controlling for English proficiency and education (Table 6), suggesting that there are other barriers to volunteering besides these two factors.

The regression results on provision of child care were similar to the descriptive results, showing the immigrant aged from Italy, Greece and the Asian countries (except Northeast Asia aside from China) more likely to be caring for children than those from Western Europe (Table 7). The regression results also confirm that differences by birthplace in caring for older, sick or disabled family members or others were small and not significant, with the exception of the immigrant aged from Italy who were more likely than other immigrant aged to be caring for other family members. The regression results also show that carers were more likely to be women and to be married, and were also more likely to be proficient in English and to have higher education. More recently arrived aged migrants were also more likely to be caring for children than immigrants who arrived before 1996.

Conclusion

The immigrant aged in Australia in 2006 differ by country or region of origin in many of the measures of social and economic well-being, even after taking into account differences in their demographic and human capital characteristics. This suggests that cultural factors and migration experiences associated with country of origin as a measure of ethnicity have some influence on the well-being of the immigrant aged.

The immigrant aged from Italy and Greece are the two largest groups of older immigrants of non-English-speaking background. Most have lived in Australia for more than 30 years. They show many distinctive characteristics. Both groups have higher proportions married and/or living at home and helping to look after grandchildren and other family members in need of care. Although the indicators of independent living show they are more likely to need assistance with daily living activities and to be in the lower income group, they are less likely to be in institutional care than the Australian-born aged. In contrast to the immigrant aged from these two countries and other Southern and Eastern European countries, those from Western Europe are more similar to the Australian-born aged and immigrants from the English-speaking countries. These findings are similar to those of earlier studies of the European-born aged (AIMA 1985, 1986; Rowland 1991).

There are also differences among the immigrant aged from the Asian countries. The immigrant aged from Malaysia and Philippines are more proficient in English, have more education and higher income than older migrants from China and Vietnam. Because of their lower English proficiency and lower income, the immigrant aged from China and Vietnam may be more dependent on their family.

The findings indicate that active retirement for the immigrant aged tends to be focused within the family rather than in the broader community. About one in five older people of Southern European and Asian background continue to make a contribution to their families’ welfare as caregivers. While it has been suggested that having a productive role in the family can give purpose and meaning to the daily lives of the aged (Holzberg 1982), the data do not inform us about the perception of older people in relation to their care-giving role. The greater stability of marriage that is likely to be related to social, cultural and religious beliefs that are associated with some of the Southern European and Asian ethnicities also means that these immigrant aged are more likely to have the support and family resource of a spouse, which is shown to facilitate living at home.

The family focus of the immigrant aged from most of the non-English-speaking countries appears to have the complementary effect of less participation in community activities such as volunteering. While it was thought initially that perhaps the lower education and lack of English of some of the immigrant aged might have inhibited their community participation, controlling for these characteristics in the data analysis did not eliminate ethnic disparities in involvement in volunteering. This may suggest the absence of a tradition of volunteering in some cultures, or less social integration among some of the immigrant aged that are related to other (perhaps cultural) factors besides lack of proficiency in English.

Of more concern is the greater likelihood of poorer physical health and lower income among some groups of the immigrant aged. Those from Southern and Eastern Europe, the Middle East and the Pacific region are particularly at risk of needing assistance with daily activities, while the lower income of the immigrant aged from some Asian countries and the Middle East suggest that they are likely to be dependent on government income support. The circumstances of all the above birthplace groups may be related to their migration or personal histories associated with country of origin rather than ethnic or cultural factors; however, it is not possible to examine this with this dataset. The poorer health and lower income among some of the immigrant aged are two areas where policy interest in their social and economic well-being may be directed, although further studies are needed to provide more information on the nature of these conditions so that any assistance and services provided will be relevant and helpful.

While the public health and welfare systems will be important to the health and economic well-being of the immigrant aged, their social well-being is much more dependent on their families and ethnic communities. This is particularly the case when they do not speak English very well or they have a tradition of strong family networks that encourage coresidence and codependence across generations. Families and ethnic communities therefore have an important role in enhancing the ageing experiences of their older members.

The current study has examined a number of indicators of social and economic well-being of the immigrant aged and shown considerable diversity in these indicators by the migrant’s country of origin. The findings support the view advanced by Holzberg (1982) that examining the influence of country of origin can be an important step towards understanding the well-being of the ethnic aged. However, the data examined are not able to illustrate how the cultural dimensions of ethnicity can facilitate older migrants’ adjustment to the ageing process. More detailed studies of the ageing experiences of specific birthplace or ethnic groups are necessary to provide the answers to this question as there is considerable diversity both between and within the geographic regional groups examined in this paper.