Temporal Variations in the Spatial Distribution of Australia s Chinese Communities

Transcription

1 Temporal Variations in the Spatial Distribution of Australia s Chinese Communities James E. Coughlan Senior Lecturer Department of Anthropology, Archaeology and Sociology School of Arts and Social Sciences Faculty of Arts, Education and Social Sciences James Cook University Cairns, Queensland, 4878, Australia (James.Coughlan@jcu.edu.au) (Phone: ) Paper presented at The Re-imaging Sociology Conference of The Australian Sociological Association, 2-5 December, 2008, University of Melbourne, Melbourne, Victoria, Australia. James Eric Coughlan, 2008.

2 Temporal Variations in the Spatial Distribution of Australia s Chinese Communities Paper presented at The Re-imaging Sociology Conference of The Australian Sociological Association, 2-5 December, 2008, University of Melbourne, Melbourne, Victoria, Australia

3 i Abstract The immigrant ethnic Chinese communities in Australia have emigrated from a diversity of countries, predominantly in the Asia-Pacific region, with the main source countries being Malaysia, the People s Republic of China, Singapore and Viet Nam, as well as Hong Kong Taiwan. In addition to these immigrant populations, there are also expanding second and later generations of Chinese- Australians. This paper describes the changing spatial distribution and concentration of Chinese communities in Australia over the two past decades, through an examination of national Statistical Local Area (SLA) level data from recent Australian Censuses. Rather than focus analysis on country of birth data, this research analyses data of individuals (i) of Chinese ancestry and (ii) who spoke a Chinese language at home. A number of SLA factors are considered in an attempt to provide an initial discussion of meso-level factors which have contributed to changes in the Chinese-Australian distributions and concentrations. The data reveal a contrasting picture of the spatial distribution and concentration of the Chinese- Australian communities, and indicate that there are few areas of high Chinese spatial concentration, and that overall there is an increasing dispersion of Chinese-Australians into the broader Australian community. Keywords: Chinese, spatial, concentration, dispersion, ancestry, language.

4 1 Introduction Over the past three decades, Australia has settled approximately 300,000 Chinese refugees and immigrants from a diversity of countries across the globe, although most of these immigrants were born in Hong Kong, Malaysia, the People s Republic of China, Singapore, Taiwan and Viet Nam. These Chinese immigrants have come from a diversity of socio-economic backgrounds, and have entered Australia under an assortment of immigration programmes. There was debate in the 1980s and into the early 2000s about the formation of Vietnamese and other Asian enclaves and ghettos 1 in some Australian cities (Birrell 1993; Healy and Birrell 2003), although most research at the time discounted such formations (Viviani 1984; Coughlan 1989b; Viviani et al. 1993; Dunn 1998: ; Johnston et al. 2001a; Poulsen et al. 2004); 2 it is time to examine more contemporary data to assess the spatial integration of Chinese-Australians into mainstream Australian society. This paper describes the changing spatial distribution and concentration of Chinese communities in Australia over the two past decades. 3 Rather than focus the analysis on merely country of birth data, as has been the practice in most previous Australian studies, 4 this research analyses data of individuals (i) of Chinese ancestry and (ii) who spoke a Chinese language - either Cantonese, Mandarin or Other Chinese languages - at home. 5 A number of basic demographic and economic factors at the SLA level are considered in an attempt to provide an initial discussion of meso-level factors which may have contributed to the observed changes in Chinese-Australian distributions and concentrations since The principal objective of this paper is to provide a clear description of the spatial distribution and concentration of Australia s Chinese communities by conducting a time-series analysis of data from four of the last five Australian Censuses. In addition to providing a purely statistical analysis - through the use of indices of isolation and segregation, 7 as well as regression analysis - which has been the sole focus of some previous research (such as Johnston et al. 2001a; 2001b), this paper seeks to also identify SLAs of high Chinese concentration and SLAs which have exhibited substantial changes in their Chinese population over the period 1986/ Theories of Spatial Concentration and Distribution

5 2 Over the past five decades or so there have been a diversity of theoretical perspectives advanced to assist in explaining the spatial concentration and distribution of ethnic groups. 8 While most of these theories have been advanced to explain ethnic spatial concentration and distribution in countries which have large immigrant populations, such as Australia, Canada, New Zealand, the United Kingdom and the United States of America, generally these theories have ignored the prevailing government philosophies/policies directed towards immigrant settlement or ethnic minorities. 9 While policies and theories of assimilation dominated in the above countries from the nineteenth century up until the 1960s, the notion of pluralism was initially advanced in the 1910s and gained broader currency in the 1960s and 1970s as being a more realistic explanation and philosophy of integrating ethnic and immigrant populations into the mainstream of society than assimilation (Zelinsky and Lee 1998: ). In the 1970s, Australia and Canada introduced policies of multiculturalism, which acknowledged and fostered cultural pluralism, rather than an enforced assimilationist position. It is against this background that we need to take into account theories of ethnic group concentration and dispersion. This paper does not seek to discuss the various theories which have been advanced to explain ethnic group concentration and dispersion, as there is already an adequate literature on this issue (Dunn 1998: ; van Kempen and Özüekren 1998: ), but rather it is noted that any theoretical propositions relating to ethnic group concentration and dispersion advanced in countries which have assimilationist policies, will probably not be applicable in explaining patterns in countries which have adopted multicultural policies towards immigrant and ethnic groups (van Kempen and Özüekren 1998: 1648). Most theories of ethnic group concentration and dispersion frequently failed to take into consideration the prevailing and changing nature of local, national and international economic relations, especially those relating to production and the work environment, and generally most contemporary theories continue this omission (van Kempen and Özüekren 1998: 1632). In addition, these theories generally fail to acknowledge the changing ethnic and ethnic composition of immigrant flows. As Zelinsky and Lee (1998: 283) note with respect to the United States of America, and equally applicable to the other countries listed above:... since 1965, a massive influx of immigrants, refugees and sojourners..., along with a significant restructuring of the world economy and society, obliges us to re-evaluate

6 3... the assimilationist perspective.... In racial appearance, persons arriving from... Asia, the Pacific Islands, Africa, the Caribbean and Latin America are seldom mistaken for Europeans. The spatial concentration of new immigrants from particular countries in a geographic area and the creation of so-called ethnic enclaves is frequently criticised by members of the host society who fear an erosion of social cohesion. However, numerous authors (for example, McKay 1981: 1-2; Fieldhouse 1999: ; Musterd et al. 1999; Johnston et al. 2001b; 2002; Farrell 2008: ) have explained that the structure of, and possible discrimination in, housing and labour markets (as well as the location of migrant hostels in the case of Australia) and the arrival of additional family members and friends, all contribute to the creation of ethnic clustering. At the same time, ethnic clustering not only assists the new immigrants in their transition into their new environment, but also permits the provision of group services, such as food, medical and social services, to assist in the process of familiarisation with, and integration into, the host society. Residential clustering is exacerbated when the new immigrant group has few skills that are readily transferable to the host society and low educational qualifications, forcing them into unskilled low income occupations, or unemployment, and therefore cheap housing (Darroch and Marston 1971: 502; Fieldhouse 1999: ; Johnston et al. 2001b). Seven decades ago Cressey, building on the previous work of the Chicago School of Sociology, developed a four stage model of immigrant group spatial distribution. 10 This model, with some minor modifications, proposes that during the first stage of settlement a few immigrants move into an area, closely followed by a large number of their compatriots (often family members). This stage is termed invasion (Cressey 1938: 62). Conflict with the host community may accompany invasion, depending on the prejudices and cultural differences of the groups concerned. Following the invasion is the recession stage, during which some of the long-time residents of the area move out as, in their eyes, the desirability of the area diminishes. Recession may, however, come first if the area is perceived as being obsolete in the eyes of the long-time residents, which in turn creates cheaper housing possibilities for the new immigrants. In this situation recession invites invasion. The third stage of the process, domination, involves a reorganisation of the social (and economic) life of the area around the institutions of the new immigrants. These social and economic institutions are initially established due to the new immigrants perceived disadvantages relative to the host society at large, and over time they become an influential force in the local community. During the final stage, which may be termed dispersion, by which time the new immigrants have become established in the

7 4 host society and improved their economic well-being, and possibly their socio-economic status, the new immigrants move out into other areas, into mainstream communities, and the process starts again with the arrival of a new ethnic or immigrant group. 11 Actual improved economic well-being may not be the only reason for dispersion, as after a time residents may perceive themselves as being less disadvantaged than their newly-arrived compatriots, and thus seek out better accommodation and a more appropriate lifestyle (Loo and Mar 1982: ). Another possible contributing factor for dispersion is if there are significant cultural or socio-economic differences between the earlier immigrants and their newly arrived compatriots. In summary, the formation of spatial concentrations of new immigrants may be perceived as a process whereby new immigrants are initially able to establish themselves in the host society, and hopefully prepare themselves, both economically and socially, for integration into the broader host society. These spatial concentrations will continue to exist over time as more immigrants from the source country continue to arrive in Australia, despite the geographic dispersion of the earlier arrivals; that is, over time spatial concentrations will continue, but at the same time the spatial dispersion also increases, as illustrated below. Chinese Spatial Distribution and Concentration To place the ensuing analysis and discussion in some perspective, Table 1 12 provides data, when collected, on the number of people of Chinese ancestry, were born in the People s Republic of China and spoke a Chinese language at home, from the past seven Censuses, and what proportion these numbers were of the total Australian population at the time. The data clearly indicate that the People s Republic of China-born population increased markedly between 1986 and as a result of the Tiananmen Square incident on 4 June 1989, and the granting of special temporary visas to Chinese nationals in Australia at the time as a result of the incident 13 - and then again between 2001 and Concurrently, there has been a sizeable increase in the number and proportion of Australians of Chinese ancestry and who speak a Chinese language at home. Clearly this latter pattern is a result of two factors: firstly, the growth in the size of the second and later generation Chinese-Australian communities, and secondly the immigration of ethnic Chinese. The data in Table 1 indicate that between 1986 and 2006, the People s Republic of China-born population increased by percent, while the Chinese ancestry population increased by per cent and the Chinese speaking population increased by per cent. As we anticipated above, these data suggest that

8 5 county of birth data alone are not going to provide a clear indication of the spatial distribution of Chinese-Australians, even if we included those born in Hong Kong, Macau and Taiwan, and thus we need to also consult ancestry and language spoken data. [Table 1 about here] Spatial Distribution of the Chinese Languages Speakers As noted at the beginning of this paper, while the country of birth data provide a good understanding of the spatial concentration and distribution of individuals born in various Chinese countries, it does not provide a reliable indication of the distribution of the Chinese communities due to the growing size of the second and later generations of Chinese-Australians and the reality that most ethnic Chinese in Australia were not born in the Hong Kong, People s Republic of China or Taiwan, although a significant minority were born in other countries in East and South-East Asia. In order to gain a better understanding of the spatial concentration and distribution of the Chinese- Australian communities, the paper commences its analysis of Chinese languages data. The middle third of Table 2 presents data which initially suggests a dispersion of the Chinesespeakers between 1996 and The data suggest a marginal increase of 0.6 per cent in the number of SLAs with Chinese language speakers in the wake of a 45.3 per cent increase in the size of the Chinese-speaking population, suggesting, at a first glance, that there has been no dispersion of Chinese-speaking Australians. The disaggregated data suggest that while the Cantonese-speakers are dispersing, the other two languages groups are increasingly congregating together, or that there has been a substantial loss of Chinese language speakers over the last decade, which the data indicate is not the case. 15 The figures in this Table are relatively meaningless unless we can identify where these changes have taken place and their magnitudes. [Table 2 about here] Another issue to be examine is the proportion of an SLA s population who speak Chinese at home. Table 3 presents data from four Censuses, listing out in order based on the 2006 Census SLA data, the proportion of each SLA s population that speaks Chinese at home, and clearly indicate that there are essentially five enclaves of Chinese language speakers in Australia, with four of these being

9 6 in Queensland - in other words, five SLAs have more than 20 per cent of their population who speak Chinese at home, 16 and, after rounding, 30 SLAs have per cent of their population who speak Chinese at home. Recalling from Table 1 that the People s Republic of China-born population increased substantially during , and with the arrival of the second and later generations, then the data suggest that Chinese language enclaves are forming in a handful of SLAs. [Table 3 about here] An examination of the data indicated that the majority of 1996 to 2006 changes involve relatively small numbers, and thus Table 4 considers the major gains and losses in SLAs which had more than 200 Chinese-speaking persons as enumerated in the 1996 or 2006 Censuses. From this Table it is clear that while there have been a few large gains in some SLAs, concurrently there have been substantial declines in SLAs which had sizeable numbers and concentrations of Chinese language speakers in [Table 4 about here] To more clearly quantify the above changes, and to verify the mixed messages, indices of isolation and segregation were calculated for the three language categories, and these are presented in Table 5. The indices in middle part of Table 5 suggest a clear increased segregation and isolation of the Chinese speakers. We do not have the data to investigate this issue further, but certainly the individual characteristics of the different Chinese languages speakers concerned needs to be investigated. [Table 5 about here] In order to understand the above changes, basic correlation analysis was conducted with a number of SLA demographic and socio-economic variables. 17 At the SLA level, the most potent Pearson Correlation coefficients (significant at the p<0.001 level) with the proportion of each SLA s population that were Chinese language speakers were identified, and to assess these relationships further, scattergrams were generated, and the results of these informed the regression analysis. The regression model for the proportion of each SLA s population who were Chinese-speakers returned significant results (R 2 = 0.457), with the model presented in Table 6, indicates a strong direct

10 7 relationship between the proportion of the SLA population born overseas (excluding the People s Republic of China-born) and the number of people per bedroom, indicating that SLAs with a high proportion of Chinese speaking residents have a high proportion of residents born overseas (excluding the People s Republic of China) and household sizes resulting in above average crowding; a strong inverse relationship exists with the proportion of the SLA s population who were born in the People s Republic of China and arrived in Australia before 1991, household size and the median age of the SLA s population, indicating that SLAs with high Chinese-speaking concentrations have a low number of People s Republic of China-born residents who settled in Australia before 1991, have smaller than average households (with a small number of bedrooms), and a more youthful age profile than average. [Table 6 about here] With this reasonably good understanding of the magnitude and nature of the spatial concentration and distribution of the Chinese languages speaking communities, the final issue for consideration in this paper is the issue of Chinese ancestry. Spatial Distribution of People of Chinese Ancestry Unfortunately, spatial ancestry data from the 1986 Census at the SLA level is not readily available, and thus the discussion here will be limited to the last two Censuses. Table 7 presents data from the last two Censuses, listing out in order based on the 2006 Census SLA data, the proportion of each SLA s population of Chinese ancestry. Similar to the Chinese-speakers data in Table 3, the data in Table 7 clearly indicate that there are essentially ten enclaves of Chinese ancestry in Australia, half of which are in Queensland - in other words, ten SLAs have more than 20 per cent of their population who are of Chinese ancestry, and, after rounding, 39 SLAs have between per cent of their population who are of Chinese ancestry. Over the period, of the 20 SLAs in Table 7, 17 recorded an increase in their proportion of Chinese ancestry and two recorded a decrease. These data, together with those from the previous section, would suggest that in SLAs with large Chinese ancestry and Chinese-speaking communities, small Chinese ancestry enclaves are forming in about a dozen SLAs. [Table 7 about here]

11 8 The bottom portion of Table 2 presents data which initially suggests a dispersion of the Chinese ancestry population between 2001 and An examination of the SLA data indicates that the majority of changes involve relatively, and thus Table 8 only considers the major gains and losses in SLAs which had more than 150 people of Chinese ancestry as enumerated in the 2001 or 2006 Censuses. From this Table it is clear that while there have been a few large gains in some SLAs, most of the SLAs which experienced large proportional gains between 2001 and 2006 had less than 1,000 residents of Chinese ancestry in In other words, the gains were predominantly in SLAs with a relatively small number of residents of Chinese ancestry. At the same time there have been only marginal declines in SLAs with sizeable numbers and concentrations of residents of Chinese ancestry in 2001, and no indication of any substantial declines. [Table 8 about here] Repeating procedures detailed previously to more clearly quantify the above changes, and to verify the contrasting patterns, indices of isolation and segregation were calculated; the indices at the bottom part of Table 5 mirror those for Chinese languages speakers by indicating an increase in isolation, but a decrease in segregation, of individuals of Chinese ancestry, which once again confirms our earlier assessment that the data suggest both increased concentration and increased dispersion of individuals of Chinese ancestry, as well as Chinese languages speakers. The regression model for the proportion of each SLA s population who were of Chinese ancestry returned significant results (R 2 = 0.513), with the model presented in Table 9 indicating a strong direct relationship between the proportion of the SLA population born overseas (excluding the People s Republic of China-born), median household income and the average number of people per bedroom, indicating that SLAs with a high proportion of residents of Chinese ancestry have a high proportion of residents born overseas, household sizes resulting in crowding and, surprisingly, high household incomes 19 ; a strong inverse relationship exists with average household size, the median age of the SLA s population and the proportion of the SLA population who were born in the People s Republic of China and arrived in Australia before [Table 9 about here]

12 9 Discussion While many of the Chinese who came to Australia in the 1960s and 1970 were Chinese overseas students from South-East Asia and Hong Kong, many of whom subsequently settled in Australia, the 1980s witnessed both an expansion of this Chinese student migration to include students from the People s Republic of China and Taiwan, as well as an ethnic Chinese refugee immigration from Indochina (Cambodia, Laos and Viet Nam). The events of June 1989 in Beijing precipitated both a refugee-type movement from the People s Republic of China, as well as an enhanced orderly emigration from this country. Thus, the latter part of the 20th Century and the early years of the 21st Century witnessed a substantial movement of citizens from the People s Republic of China to Australia, and other developed Western economies. The 1980s and 1990s evolution of numerous Chinese spatial concentrations in the Brisbane, Melbourne and Sydney metropolitan areas was primarily due to a saturation effect. Low cost housing in areas in close proximity to migrant hostels were quickly taken up by the first wave of ethnic Chinese refugees from the Indochinese countries. As the cheap housing available in these areas dried up, other low cost housing areas, close to (possible) places of employment and public transport were sought by the more recently arrived refugees from the People s Republic of China. These areas soon attracted more recently arrived Chinese refugees and immigrants, thus producing additional locations of Chinese concentration. At the same time, the relatively cheap government housing sector was exploited by the Chinese from Indochina, as well as other ethnic groups from Indochina (Birrell 1985: 37-40). After a waiting period of generally several years, some Chinese-Indochinese moved into government housing, often in government housing estates, if it was perceived to be to their economic benefit, thus resulting in newer areas of Chinese concentration appearing in most State/Territory capital cities. The analysis of the 1986 to 2006 Australian Census data presented above, suggest that at least spatially most Chinese-Australians are integrating into mainstream Australian society, to a degree. The above analysis of ancestry and language spoken at home data has allowed us to quantify, to a degree, the magnitude of this spatial integration and segregation, with the analysis also identifying the presence of about a dozen or so moderately sized Chinese enclaves. However, we must not forget that small Chinese, Greek, Italian, etc. enclaves have persisted in some Australian capitals since at least the middle of the 20th century, and these enclaves perform important economic and social

13 10 functions. Since the early 1990s, several factors have contributed to the spatial concentration and dispersion of Chinese-Australians. Firstly, the nature and source countries of Chinese immigration changed from the predominantly South-East Asian international student and Indochinese refugee inflow of the 1970s and 1980s, to the added component of the People s Republic of China refugee component in the 1990s, to a family reunion migration direct from various Asian countries in the 1990s and 2000s. Thus, most of the Chinese who settled in Australia over the past two decades have moved straight into the community, having joined immediate family members here. Secondly, from a human capital perspective, the nature of the community has changed substantially. The former international student component, which has been present since the 1960s, have acquired tertiary qualifications from Australian educational institutions, and this component continues to be significant up until the present time. On the other hand, the longer-term Chinese immigrants who settled in Australia under refugee or family reunion programmes, the so-called humanitarian programmes, have acquired more Australian work experience, which has allowed them to obtain secure and higher paying employment. At the same time, most younger members of the community have been educated in Australian schools, and a high proportion have entered tertiary education institutions. This acquisition of Australian educational qualifications, and well as greater proficiency in English than their parents, permitted these younger Chinese to enter the mainstream labour force, rather than the segmented marginal labour forces where proficiency in spoken English is not required, with a high proportion entering professional occupations (Coughlan 1991; 1992b; 1994a; 1994b; 1999; Coughlan and Ho 1997). The job security and high incomes such occupations attract, permitted these younger Chinese to purchase their own homes, and to move into middle class Australian suburbs, often bringing their ageing parents with them. 21 Therefore, when we examine the detailed spatial data, we find a small Chinese-Australian presence in most urban SLAs across Australia. Conclusions and Further Research It is critically important to acknowledge that ethnic segregation may be brought about due to a diversity of factors, such as (i) the desire of some members of ethnic groups to live in close proximity to their co-ethnics, for a variety of reasons, (ii) enforced societal structural discrimination, which often causes (iii) disadvantage and poverty.

14 11 The spatial concentration and distribution the Chinese-Australian communities has been neither the subject of public debate nor concern, in spite of such concerns having been expressed for the Vietnamese communities in Australia, an ethnic group which is substantially smaller than the ethnic Chinese communities in Australia. We can find only two studies (Johnston et al. 2001a; 2001b) in recent times which have addressed the issue of the spatial concentration and distribution of Chinese-Australian communities, and their findings are similar to those reported here. Most Chinese who have steady employment and are purchasing homes move away from the less well-off Chinese enclaves. As the analysis of the previous sections has shown, the Chinese are gradually dispersing, and do not all live in a dozen or so Chinese enclaves, and although the 2006 Census data clearly indicate that such moderately sized enclaves clearly exist, these enclaves are home to only a minute proportion of the entire Australian-Chinese population. Our analysis of ancestry and language spoken at home data at the SLA level from the 1986 to 2006 Censuses reveals the presence of about a dozen moderately sized Chinese enclaves, findings similar to those reported by Johnston et al. (2001a: 154; 2001b: 584) 22 in their CD-based analysis of 1996 Census data in Sydney, and in support of Cressey s (1938) four stage model of immigrant group spatial distribution. Indeed, unlike most previous studies of Chinese-Australian spatial distribution and concentration, which have just focussed on one or two cities, we have taken a national approach, and found that Chinese-Australians of all generations are generally spatially dispersing into multicultural Australia, while at the same time there are signs of increased spatial segregation. Our analysis identified two meso-level SLA factors which appear to be contributing to areas of high concentration of Chinese-Australians - high proportion of overseas-born (excluding the People s Republic of China-born) and period of residence in Australia - across the two ethnicityrelated variables considered in this paper. This paper has attempted a first crude time-series analysis of census data into the spatial distribution and concentration of Chinese-Australians. Additional analysis needs to be conducted at the CD level, using the ethnicity-related variables used here. Undoubtedly, such an analysis will identity a small number of capital city neighbourhoods of high Chinese-Australian concentration, in other words, neighbourhood enclaves. In addition, further analysis needs to determine the culturaldemographic and human capital attributes of Chinese-Australians who reside in areas of high Chinese

15 12 concentration and those who do not. An analysis of population census internal migration data should be conducted to ascertain the characteristics of Chinese-Australians who have moved out of, or in to, areas of high concentration.

16 13 Endnotes 1. See Peach (1996: ) for a critical discussion of this concept. There seems to be little agreement in the literature on what level of ethnic concentration constitutes an enclave. Nevertheless, there does seems to be some consensus that where at least per cent of a spatial location is inhabited by one particular ethnic minority group, then that area may be considered an ethnic enclave. 2. Johnston et al. (2001a: 154) acknowledge in their Collection District analysis of 1996 Census data for the Viet Nam-born in Sydney, that the data indicate... the existence of Vietnamese enclaves, but an absence of ghetto-like concentrations. 3. Where appropriate, in the following tables, situations where SLA boundaries have changed between the time of the 1986 and 2006 Censuses are noted. As Farrell (2008: 476) notes In any longitudinal study of segregation, the compatibility of tract [spatial] boundaries is a salient issue. Failing to account for boundary changes - these commonly occur between census enumerations due to population change - would bias any conclusions about changing patterns of segregation in a given metropolitan area.. 4. Burnley (1994) appears to have been the first, and Poulsen et al. (2004) the second, to date, Australian researchers to use ancestry data when analysing the residential concentrations of ethnic communities. 5. The 2006 Census questions were Does the person speak a language other than English at home? (with the instruction If more than one language other than English, write the one that is spoken most often. ) and What is the person s ancestry? (with the instruction Provide up to two ancestries only. ). Where applicable, previous censuses sought the same information, although there were slight variations in question wording and question instructions. 6. Farrell (2008: 468) advises that One must also take into account how meso-level community distinctions factor into metropolitan segregation. (emphasis in original). 7. The paper utilises the Index of Isolation developed by Lieberson (Johnston et al. 2005: ): II = n i=1 [(x i /X) * (x i /t i )] where x i X t i i n = number of members of the ethnic group in geographical area i (SLA), = total population of the ethnic group (nationally), = total population of the geographical area i (SLA), = one of the n geographical areas, and = number of geographical areas,

17 with values ranging from 0 (no segregation) up to 1 (total segregation); and an Index of Segregation used by Musterd (2003: 639): 14 IS = 0.5 n i=1 [(x i /X) - (y i /Y)] * 100 where x i X y i Y i n = number of members of the ethnic group in geographical area i (SLA), = total population of the ethnic group (nationally), = number of the rest of the population in geographical area i (SLA), = total number of the rest of the population (nationally), = one of the n geographical areas, and = number of geographical areas, with values ranging from 0 (no segregation) up to 100 (total segregation). There has been much debate over the utility of the various measures of segregation (see for example, Johnston et al. 2005; Massey and Denton 1988; White 1983), and their delivering contradictory results, as will be seen below. The two measures used in this paper provide some quantitative indication of dispersion and segregation, and they are used cautiously and reluctantly. 8. See Dunn (1998: ) and van Kempen and Özüekren (1998: ) for a brief critical discussion of this literature. 9. Dunn (1998), Johnston et al. (2002) and Poulsen et al. (2004) are amongst the few exceptions here. 10. For alternative theories advanced around this time, see works by Park, Wirth and Burgess, Warner, and Srole cited in Loo and Mar (1982) and Farrell (2008: ). 11. See Ford (1950) and Taeuber and Taeuber (1964) for additional discussion of this model. 12. More detailed versions of most tables presented in this paper are available from the author upon request. 13. The ramifications of the incident on the situation of the People s Republic of China s citizens in Australia at the time of the incident, and its subsequent impact on the Chinese communities in Australia, is discussed in Coughlan and Ho (1997) and Coughlan (1999). 14. Of the SLAs which experienced growth in their number/proportion of Chinese languages speaking residents, 85 (6.0 per cent) SLAs had no Chinese languages speaking residents in 1996, while of those which experienced a decline in their number/proportion, 78 (5.5 per cent) SLAs had Chinese languages speaking residents in 1996 but none in The 1996 (2006) Census enumerated 202,494 (244,549) Cantonese speakers, a 20.8 per cent increase from 1996 to 2006; 92,360 (220,600) Mandarin speakers, a per cent increase; and, 49,465 (35,318) Other Chinese languages speakers, a 28.6 percent decline. These data, examined with the SLA data, would suggest that Cantonese speakers are dispersing, while Mandarin speakers are increasing in concentration.

18 15 Whether the decline in the number of Other Chinese languages speakers is real, due to emigration or mortality, or is a result of language loss, or is an artefact of the situation where many individual ethnic Chinese born in South East Asian countries are fluent in a number of Chinese dialects and languages, and some of these individuals may have changed their census responses between the 1996 and 2006 Censuses, is unknown and requires further analysis. 16. However, in most of these SLAs, the number of Chinese speakers is not large - Territory of Christmas Island 471, Robertson 1,495, Stretton-Karawatha 984, MacGregor 1,240 and Sunnybank 1,849. However, the numbers in Hurstville City (at 19.2 per cent, and 14,182 speakers) and Auburn Area (at 18.3 per cent and 11,997 speakers) may be considered high. 17. These factors were median age of persons in the SLA, median household income in dollars per week, median housing loan repayments in dollars per month, median housing rental in dollars per week, average number of person per bedroom in the household, average household size, SLA unemployment rate, SLA labour force participation rate, and the proportion of the SLA that was born overseas (excluding the People s Republic of Chinaborn). 18. Of the SLAs which experienced growth in their number/proportion of residents of Chinese ancestry, 46 (3.2 per cent) SLAs had no residents of Chinese ancestry in 2001, while of those which experienced a decline in their number/proportion, 49 (3.4 per cent) SLAs had residents of Chinese ancestry in 2001 but none in Although the latter two factors may be related, the data did not indicate a multi-collinearity issue. 20. It may be recalled that the same patterns were present for the analysis of the Chinese language speakers. 21. These issues are briefly noted in Farrell (2008: 472). 22. Our analysis of the 2006 Census data also found that period of residence in Australia was as an important factor. Considering that our analysis was conducted with 2006, rather than 1996, Census data, it would seem that period of residence in Australia is still a very important explanator of Chinese spatial concentration over at least the past decade.

19 16 Bibliography Birrell, B. (1993) Ethnic concentrations: the Vietnamese experience, People and Place 1 (1): Birrell, R. (1985) Immigration and Social Planning, Journal of the Australian Population Association 2 (1): Burnley, I. H. (1994) Immigration, ancestry and residence in Sydney, Australian Geographical Studies 32 (1): Coughlan, J. E. (1991) A Demographic and Socio-economic Profile of Six Ethnic-Chinese Asianborn Communities in Australia: A Brief Overview From the 1986 Census, paper presented at the 12th International Conference of Historians of Asia, June 24-28, 1991, Hong Kong. (1992a) (ed.) The Diverse Asians: A Profile of Six Asian Communities In Australia, Australia- Asia Papers. Nathan: Centre for the Study of Australia-Asia Relations, Griffith University. (1992b) A Comparative Profile of Overseas-Born Ethnic Chinese Communities in Australia, Journal of Overseas Chinese Studies 2: (1994a) Community Profiles: 1991 Census - China born. Canberra: Australian Government Publishing Service. (1994b) Chinese in Australia: Chinese Community Profiles, Access China 16: (1996) The Transformation in the Characteristics of Chinese Migration to Australia, , Asian Profile. 24 (2): (1988) The Changing Characteristics of Chinese Migrants to Australia During the 1980s and Early 1990s, pp in E. Sinn (ed.) The Last Half Century of Overseas Chinese. Hong Kong: Hong Kong University Press.

20 17 (1999) The Chinese in Australia: A Socio-Economic Profile Based on the 1996 Census, paper presented at Chinese Identities - The Sixth Biennial Conference of the Chinese Studies Association of Australia - Conference, 8-10 July 1999, Murdoch University, Perth, Western Australia. (2008) The Changing Spatial Distribution and Concentration of Australia s Chinese Communities: Transformations Over the Past Twenty Years - An Analysis of Australian Population Census Data, paper presented at the 14th Bienniel Australian Population Association Conference, 30 June - 3 July, 2008, Alice Springs, Northern Territory. Coughlan, J. E. and Ho Chooi Hon (1997) The Chinese in Australia: Immigrants from the People s Republic of China, Malaysia, Singapore, Taiwan, Hong Kong and Macau, pp in J. E. Coughlan and D. J. McNamara (eds.) Asians in Australia: Patterns of Migration and Settlement. South Melbourne: Macmillan Education Australia. Coughlan, J. E. and D. J. McNamara (eds.) (1997) Asians in Australia: Patterns of Migration and Settlement. South Melbourne: Macmillan Education Australia. Cressey, P. F. (1938) Population Succession in Chicago: , American Journal of Sociology 44 (1): Darroch, A. G. and W. G. Marston (1971) The Social Class Basis of Ethnic Residential Segregation: The Canadian Case, American Journal of Sociology 77 (3): Dunn, K. M. (1998) Rethinking Ethnic Concentrations: Cabramatta, The Case of Sydney, Urban Studies 35 (3): Farrell, C. R. (2008) Bifurcation, Fragmentation or Integration: The Racial and Geographical Structure of US Metropolitan Segregation, , Urban Studies 45 (3): Fieldhouse, E. A. (1999) Ethnic Minority Unemployment and Spatial Mismatch: The Case of London, Urban Studies 36 (9):

21 18 Ford, R. G. (1950) Population Succession in Chicago, American Journal of Sociology 56 (2): Healy, E. and B. Birrell (2003) Metropolis Divided: The Political Dynamic of Spatial Inequality and Migrant Settlement in Sydney, People and Place 11 (2): Johnston, R., J. Forrest and M. Poulsen (2001a) Sydney s Ethnic Geography: new approaches to analysing patterns of residential concentration, Australian Geographer 32 (2): (2001b) The Geography of an EthniCity: Residential Segregation of Birthplace and Language Groups in Sydney, 1996, Housing Studies 16 (5): (2002) The ethnic geography of EthniCities: The "American model" and residential concentration in London, Ethnicities 2 (2): Johnston, R., M. Poulsen and J. Forrest (2005) On the Measurement and Meaning of Residential Segregation: A Response to Simpson, Urban Studies 42 (7): Loo, C. and D. Mar (1982) Desired Residential Mobility in a Low Income Ethnic Community: A Case Study of Chinatown, Journal of Social Issues 38 (3): Massey, D. S. and N. A. Denton (1988) The Dimensions of Residential Segregation, Social Forces 67 (2): McKay, J. (1981) Ethnic Communities in Melbourne: An Atlas of the Overseas Born Population. Clayton: Department of Geography, Monash University. Musterd, S. (2003) Segregation and integration: a contested relationship, Journal of Ethnic and Migration Studies 29 (4): Musterd, S., H. Priemus and R. van Kempen (1999) Towards Undivided Cities: The Potential of Economic Revitalisation and Housing Redifferentiation, Housing Studies 14 (5):

22 19 Peach, C. (1996) Does Britain have ghettos?, Transactions of the Institute of British Geographers, New Series 21 (1): Poulsen, M., R. Johnston and J. Forrest (2004) Is Sydney a Divided City Ethnically?, Australian Geographical Studies 42 (3): Taeuber, K. E. and A. F. Taeuber (1964) The Negro as an Immigrant Group: Recent Trends in Racial and Ethnic Segregation, American Journal of Sociology 69 (4): van Kempen, R. and A. S. Özüekren (1998) Ethnic Segregation in Cities: New Forms and Explanations in a Dynamic World, Urban Studies 35 (10): Viviani, N. (1984) The Long Journey: Vietnamese Migration and Settlement in Australia. Carlton: Melbourne University Press. (1996) The Indochinese in Australia , from burnt boats to barbecues. Melbourne: Oxford University Press. Viviani, N., J. E. Coughlan and T. Rowland (1993) Indochinese in Australia: The Issues of Employment and Residential Concentration (Report to the Bureau of Immigration Research). Canberra: Australian Government Publishing Service. White, M. J. (1983) The Measurement of Spatial Segregation, The American Journal of Sociology 88 (5): Zelinsky, W. and B. A. Lee (1998) Heterolocalism: An Alternative Model of the Sociospatial Behavior of Immigrant Ethnic Communities, International Journal of Population Geography 4 (4):

23 20 Table 1 Number (and Proportion of Australia s Total Population (per cent)) of Chinese by Ancestry, Birthplace and Language Counts Censuses Census Year Ancestry - Chinese Country of Birth - People s Republic of China ,542 (0.137) ,883 (0.178) ,165 36,595 (1.289) (0.235) ,882 (0.462) ,009 (0.620) , ,780 (2.965) (0.753) , ,589 (3.374) (1.030) Language Spoken at Home - Chinese Languages 127, , , , ,467 (0.888) (1.588) (1.924) (2.115) (2.495) Source: Australian Bureau of Statistics, various censuses.

24 21 Table / Changes in the Number and Proportion * of SLAs with Chinese Residents based on Chinese Ancestry, Birthplace and Language Counts , 2001 and 2006 Censuses Number (and Per Cent) * of SLAs Which Experienced a Change in Their Number/Proportion of Residents Between 1996 and 2006 with Specified Characteristics People s Republic of China-Born Number 781 (54.8) Proportion 770 (54.0) Chinese Languages Number 780 (54.7) Proportion 731 (51.3) Chinese Ancestry ( change) Number 875 (61.4) Proportion 833 (58.4) Increase No Change Decrease No Residents in Both Censuses (4.6) (2.5) (3.6) (16.4) 345 (21.8) (25.3) 250 (31.2) (27.2) 112 (33.7) (24.2) (17.5) (7.9) Source: Australian Bureau of Statistics, 2006 Census DataPacks, Time Series Profile Release 2 (ABS Cat. No ). * Percentages may not total per cent due to rounding.

25 22 Table 3 Proportion of Statistical Local Areas Population Who Are Chinese Language Speakers , 1996, 2001 and 2006 Censuses Statistical Local Area 2006 Census 2001 Census 1996 Census 1986 Census Territory of Christmas Island Robertson - (Qld) Stretton-Karawatha - (Qld) MacGregor - (Qld) Sunnybank - (Qld) Burwood (A) - (NSW) Hurstville (C) - (NSW) Sunnybank Hills - (Qld) Auburn (A) - (NSW) Eight Mile Plains - (Qld) Acton - (ACT) Kogarah (A) - (NSW) Strathfield (A) - (NSW) Runcorn - (Qld) Baulkham Hills (A) - South - (NSW) Parramatta (C) - North-East - (NSW) Ashfield (A) - (NSW) Manningham (C) - West - (Vic) Calamvale - (Qld) Ryde (C) - (NSW) Source: Australian Bureau of Statistics, 2006 Census DataPacks, Time Series Profile Release 2 (ABS Cat. No ) and 1996 Census CLib96 Final Release (ABS Cat. No ). 1 In 1986, Baulkham Hills Area SLA comprised the current three SLAs of Baulkham Hills Area - Central, Baulkham Hills Area - North and Baulkham Hills Area - South. The 1986 Census count for this SLA was 1,537 speakers, and 5,898 in In 1986, Parramatta City SLA comprised the current four SLAs of Parramatta City - Inner, Parramatta City - North-East, Parramatta City - North-West and Parramatta City - South. The 1986 Census count for this SLA was 1,861 speakers, and 8,121 in 1996.

26 23 Table 4 Number and Percentage Change in the Number and Proportion of Chinese Language Speakers in Statistical Local Areas Between 1996 and and 2006 Censuses Statistical Local Area Chinesespeakers Census Chinesespeakers Census Percentage Change in the Number of Chinese-speakers in SLA Between 1996 and 2006 Censuses Percentage Change in the Proportion of Chinese-speakers in SLA Between 1996 and 2006 Censuses Gungahlin - (ACT) Wyndham (C) - South - (Vic) , Melton (S) - East - (Vic) , Molendinar - (Qld) , , Murray Bridge (RC) - (SA) , , Melbourne (C) - Inner - (Vic) 225 2,661 1, Melbourne (C) - Southbank , Docklands - (Vic) [Brisbane] City - Remainder (Qld) Kuraby - (Qld) Salisbury (C) - Balance - (SA) Port Phillip (C) - St Kilda - (Vic) 1, Middle Park - (Qld) Darebin (C) - Northcote - (Vic) 1,666 1, New Farm - (Qld) Cleveland - (Qld) Bathurst Regional (A) - Part A (NSW) Latrobe (C) - Morwell - (Vic) Marrickville (A) - (NSW) 4,291 2, Yarra (C) - Richmond - (Vic) 2,323 1, Territory of Christmas Island Source: Australian Bureau of Statistics, 2006 Census DataPacks, Time Series Profile Release 2 (ABS Cat. No ).

27 24 Table Statistical Local Area Indices of Isolation and Segregation for Chinese Residents based on Chinese Ancestry, People s Republic of China Birthplace and Chinese Language Counts , 2001 and 2006 Censuses Indices and Characteristics 1996 Census 2001 Census 2006 Census People s Republic of China-Born Index of Isolation Index of Segregation Chinese Languages Index of Isolation Index of Segregation Chinese Ancestry Index of Isolation Index of Segregation Source: Australian Bureau of Statistics, 2006 Census DataPacks, Time Series Profile Release 2 (ABS Cat. No ).

28 25 Table 6 Regression Analysis on the Proportion of SLA Population Who Speak Chinese Languages at Home Census Variables B Standard Error Significance (p) (Constant) Median Age Median Household Income (dollars per week) Median Housing Loan Repayments (dollars per month) Median Housing Rental (dollars per week) Average Number of People per Bedroom in Household Average Household Size (persons) SLA Unemployment Rate SLA Labour Force Participation Rate Percentage of SLA Born Overseas (Excluding the People s Republic of China-Born) Percentage of the People s Republic of China-Born who Arrived in Australia in the Period before R 2 = F = P < Source: Australian Bureau of Statistics, 2006 Census DataPacks, Time Series Profile Release 2 (ABS Cat. No ).