Australian Institute of Criminology

Skip to content

Migrant sex worker demographics

Overall, 70 percent (n=412) of respondents were classified as migrants and 25 percent (n=151) were classified as non-migrants. The rest of the respondents (5%, n=29) were categorised as ‘missing’ migrant status. Migrant status was coded as missing for cases where the respondent did not answer the survey question on birth country, or if their answer for birth country did not correspond with other key survey questions on migration. For example, a case was dropped if the respondent identified that they were born in a country other than Australia but selected the options ‘I have only lived in Australia’, ‘I have always lived in Australia’ or ‘I was born in Australia’ in subsequent questions.

Of the identified migrant survey population, 44 percent indicated they were born in Thailand, 26 percent in China, nine percent in South Korea and five percent in New Zealand (Figure 1). Thai, Chinese and Korean-speaking migrants were, as described previously, specifically targeted for survey collection. The majority of respondents from Thailand (87%), China (59%) and New Zealand (57%), and all respondents from South Korea, regarded their birth country as their home country (Table 3). Those born in ‘other’ category countries were more likely to consider Australia as their home country.

Figure 1 Birth country of migrant respondents (%)

N=412

a: ‘Other’ birth countries included (in alphabetical order) Brazil, Colombia, Costa Rica, England, Fiji Islands, Germany, Greece, Hong Kong, Hungary, India, Indonesia, Ireland, Japan, Kenya, Malaysia, Papua New Guinea, Persia (Iran), Philippines, Russia, Scotland, Singapore, South Africa, Spain, Sweden, Taiwan, United Kingdom, United States of America, Vietnam, Wales and Zimbabwe. In some cases, regions instead of countries were given as responses, such as Europe. Those respondents who selected the ‘other’ category for birth country but failed to specify which country (n=3) were counted as ‘other’ birth country

Source: AIC, Sex Worker Migration and Vulnerabilities to Trafficking [computer file]

Table 3 Migrant respondents’ birth country by home country (%)
Birth country
Home country New Zealand Chinaa Thailandb South Koreac Other
Australia 45 39 11 0 57
New Zealand 55 0 1 0 0
China 0 59 1 0 2
Thailand 0 1 87 0 0
South Korea 0 1 0 100 0
Other 0 0 0 0 42
Total (n) 22 100 178 37 65

a: Excludes 7 migrant respondents born in China who did not respond to the question on home country

b: Excludes 2 migrant respondents born in Thailand who did not respond to the question on home country

c: Excludes one migrant respondent born in South Korea who did not respond to the question on home country

Note: Percentages may not add to 100 due to rounding

Source: AIC, Sex Worker Migration and Vulnerabilities to Trafficking 2010 [computer file]

It should be noted that the non-migrant category included a small number of cases where respondents indicated that they were born in Australia but also answered questions regarding migration and previous country of residence. In some cases, respondents clarified that they were born in Australia but had since moved and lived overseas for a period before returning; however, there were four cases where the reasons for their responses to the migration questions were unclear. These cases were retained as non-migrants as it was not possible to verify whether the respondents were simply responding to the fact that they had returned to Australia after spending time overseas. The majority of respondents born in Australia did not respond to and/or correctly skipped the questions on migration.

Questions on age, education level, English proficiency, income expenditures and whether respondents have young children aimed to highlight the financial responsibilities of migrant respondents and potential barriers to their occupational mobility. Responses were compared both between and within migrant groups.

Gender and age

Most respondents were female (97% of migrants and 93% of non-migrants), with only 17 male and eight transgender respondents. Ten males were non-migrants, six were migrants and one was of an unclassified migrant status. All eight transgender respondents were migrants. Given the small number of transgender and male respondents, no gender comparisons were undertaken for this report.

Both migrant and non-migrant respondents had the same median age range of 30–34 years, approximating the median age of 31 years in the LASH survey and 29 years from the SSHC data (Donovan et al. 2012). However, the distribution of age ranges for migrants and non-migrants varied significantly (Figure 2). An inverse age distribution between migrants and non-migrants was observed, with non-migrants peaking in the 15–24 and 35–39 age ranges, while migrants peaked in the 25–29 and 30–34 age ranges. Respondents born in South Korea appeared to be younger than respondents born in China and Thailand (Figure 3).

Figure 2 Age group by migrant status (%)

Migrant: n=411

Non-migrant: n=150

χ(6)=39.34, p<0.001

a: Excludes one migrant respondent who did not respond to this question

b: Excludes one non-migrant respondent who did not respond to this question

Note: 7 respondents were aged 15–19 years: 3 were migrants and 4 were non-migrants. This age category was combined with the 20–24 category due to small numbers

Source: AIC, Sex Worker Migration and Vulnerabilities to Trafficking 2010 [computer file]

Figure 3 Age group of migrant respondents by birth country (%)

China: n=107

Thailand: n=179

South Korea: n=38

χ(12)=26.76, p<0.01, four cells with expected frequency <5

a: Excludes one migrant respondent born in Thailand who did not respond to this question

Note: Excludes migrant respondents born in New Zealand or ‘other’ countries due to small frequencies

Source: AIC, Sex Worker Migration and Vulnerabilities to Trafficking 2010 [computer file]

Relationship status and children

A significant difference in relationship status existed between migrant and non-migrant respondents. The majority of migrant and non-migrant sex workers were single (49% and 56% respectively; Table 4). Nineteen percent of migrant and 13 of non-migrant respondents were separated, widowed or divorced. However, among sex workers who were in a relationship at the time of the survey, migrants were significantly more likely to be married than in a de facto relationship, with the converse true for non-migrants.

More than half (55%) of migrant respondents who answered the question on children indicated they had at least one child; more than one-third (37%) had at least one child under the age of 15 years. These were similar to the proportions seen among non-migrant respondents who answered the question; however, migrant respondents were significantly more likely to have only one child (of any age) and less likely to have three or more children (Table 5). This difference was also seen in the distribution of the number of children aged under 15 years (Table 5).

Table 4 Relationship status by migrant status (%)
Relationship status Migranta Non-migrantb
Single 49 56
Registered married 20* 6*
De facto 11* 22*
Widowed 2 3
Divorced 12 7
Separated but not divorced 5 3
In a relationshipc 1 3
Total (n) 397 148

χ(6)=28.50, p<0.001, two cells with expected frequency <5

* adj res outside +/–1.96

a: Excludes 15 migrant respondents who did not respond to this question

b: Excludes 2 non-migrant respondents who did not respond to this question, and one non-migrant respondent who selected ‘other category’ for this question but did not specify whether they were in a relationship

c: Includes respondents who specified as a written response to the ‘other’ category that they had a boyfriend, partner or were in a relationship

Note: Percentages may not add to 100 due to rounding

Source: AIC, Sex Worker Migration and Vulnerabilities to Trafficking 2010 [computer file]

Table 5 Number of children by migrant status (%)
Dependent (14 years or under) Total (all ages)
Migranta Non-migrantb Migrantc Non-migrantd
None 63 66 45 49
One 28* 18* 33* 21*
Two 9 12 17 16
Three or more 0*e 4* 5* 13*
Total (n) 232 140 286 142
Significance testingf χ(3)=10.01, p<0.02,
two cells with expected frequency<5
χ(3)=14.58, p<0.01

* adj res outside +/–1.96

a: Excludes 180 migrant respondents who did not respond to this question

b: Excludes 11 non-migrant respondents who did not respond to this question

c: Excludes 126 migrant respondents who did not respond to this question

d: Excludes 9 non-migrant respondents who did not respond to this question

e: n=1, rounded to 0 percent

f: Migrant respondents were significantly less likely to answer both these questions than non-migrants; therefore, the responses may not be representative of the migrant sample and comparisons between migrant status may not be valid

Note: Percentages may not add to 100 due to rounding. The questions on children of any age and dependent age had a high non-response rate (24% and 34% respectively)

Source: AIC, Sex Worker Migration and Vulnerabilities to Trafficking 2010 [computer file]

Single mothers

Of the female migrant respondents who answered questions on relationship status and children (n=359), one-quarter were single mothers (ie they were single, widowed, divorced or separated and had at least one dependent child). Migrant and non-migrant female respondents were equally likely to be single mothers (25% cf 24%). There was a disproportionate number of missing responses from migrant respondents to the question on children; therefore the findings presented here may not be representative of the entire migrant sex worker sample and comparisons between migrant status may not be valid.

Language

Just less than two-thirds of migrant respondents who answered the question on English proficiency said they spoke English very well or well, although this figure includes respondents who were born in New Zealand and other English-speaking countries (Table 6). Nearly one-third disclosed that they did not speak English well; however, only one percent stated that they did not speak English at all. Migrant respondents born in South Korea and China were less likely to state that they speak English ‘very well’ or ‘well’ compared with Thai-born respondents.

The majority of migrant respondents indicated that they spoke English at work (Table 7); however, even these respondents did not necessarily rate their English proficiency highly. In fact, 29 percent of migrant respondents who stated that they spoke English at work also stated that they did not speak English well or at all. The languages spoken at work broadly reflected the cultural backgrounds of migrant respondents, with Thai and Chinese dialects (ie Mandarin and Cantonese) making up the majority of languages other than English spoken at work.

Table 6 How well migrant respondents spoke English, by birth country (%)
Chinaa Thailandb South Koreac Other New Zealand Totald
Very well 6 3 3 68 95 19
Well 35 70 24 25 5 46
Not well 57 27 70 8 0 34
Don’t speak English 3 0 3 0 0 1
Total (n) 104 179 37 65 22 407

a: Excludes 3 migrant respondents born in China who did not respond to this question

b: Excludes one migrant respondent born in Thailand who did not respond to this question

c: Excludes one migrant respondent born in South Korea who did not respond to this question

d: Excludes 5 migrant respondents who did not respond to this question

Note: Percentages may not add to 100 due to rounding

Source: AIC, Sex Worker Migration and Vulnerabilities to Trafficking 2010 [computer file]

Table 7 Languages spoken at work by migrant respondents (%)a
English 89
Thai 29
Mandarin 13
Cantonese 6
Korean 4
Otherb 4
Total (n) 407

a: Excludes 5 migrant respondents who did not respond to this question

b: Other languages spoken at work, where specified, included (in alphabetical order) German, Hindi, Italian, Japanese, Lao, Malay, Portuguese, Spanish, Spanish Filipino, Tagalog and Tahitian French

Note: Respondents could select multiple responses to the question on language spoken at work

Source: AIC, Sex Worker Migration and Vulnerabilities to Trafficking 2010 [computer file]

Education level

On an aggregate level (combining migrant, non-migrant and unclassified migrant status responses) education levels largely reflected those recorded in the Queensland survey on sex workers and those of Sydney respondents in the LASH survey (Donovan et al. 2012; Woodward et al. 2004). The education levels of migrant respondents to this survey, particularly those born in Thailand and China, also broadly reflect those measured among Asian-born sex workers in the SSHS (Pell et al. 2006).

Overall, migrant respondents were significantly less likely to have tertiary qualifications and more likely to have primary school level or no education compared with non-migrant respondents (χ2(2)=13.77, p<0.01). There was a corresponding significant difference in the distribution of highest education level attained between migrants and non-migrants (Table 8). Migrants were significantly more likely than non-migrants to have a bachelor degree as their highest qualification, but less likely to have a postgraduate degree, graduate diploma, certificate or Year 10. Migrants were also more likely than non-migrants to have finished Year 12 but also significantly more likely to have primary school level or no education. Thai-born respondents were particularly likely to have not gone beyond primary school education.

This demonstrates that Australian-born respondents were more likely than migrant respondents to go on to achieve tertiary-level qualifications if they completed Year 12. Similarly, Australian-born respondents who achieved a bachelor degree were more likely to go on to complete postgraduate qualifications. Closer analysis of education level by birth country shows that respondents born in New Zealand were more likely to achieve a tertiary qualification on completion of Year 12 than migrant respondents born in China, Thailand or South Korea, but nearly as likely as these groups to achieve a postgraduate qualification on completion of an undergraduate qualification.

Perhaps not surprisingly, closer analysis of education and self-rated English proficiency in the survey showed that higher-level education was associated with increased English proficiency. The distribution of English proficiency varied significantly between education levels (Figure 4).

Table 8 Education level by migrant status and birth country (%)
Migrant status Birth country (migrants)a
Migrantb Non-migrantc New Zealand Chinad Thailande South Korea
Postgraduate degree 3* 7* 0 1 2 3
Graduate diploma or certificate 5* 11* 9 1 6 3
Bachelor degree 23* 9* 23 15 28 32
Diploma 12 13 18 11 7 18
Certificate 8* 25* 27 3 6 3
High school to Year 12 27* 11* 18 41 24 37
High school to Year 10 12* 21* 5 24 9 5
Primary school to none 10* 3* 0 5 18 0
Total (n) 402 150 22 101 176 38

* adj res outside +/–1.96

χ(7)=70.02, p<0.001

a: Significance testing could not be undertaken on birth country due to low frequencies. Excludes migrant respondents who were born in ‘other’ countries

b: Excludes 10 migrant respondents who did not answer this question

c: Excludes one non-migrant respondent who did not answer this question

d: Excludes 6 migrant respondents born in China who did not answer this question

e: Excludes 4 migrant respondents born in Thailand who did not answer this question

Note: Percentages may not add to 100 due to rounding

Source: AIC, Sex Worker Migration and Vulnerabilities to Trafficking 2010 [computer file]

Figure 4 How well respondents speak English (self-rated) by education level (%)a

Tertiary: n=319

Secondary: n=213

Primary level or none: n=44

χ(4)=68.34, p<0.001

a: Includes respondents with unclassified migrant status

b: Excludes 3 respondents with tertiary education level who did not respond to the question on English proficiency

c: Excludes 2 respondents with secondary education level who did not respond to the question on English proficiency

Source: AIC Sex Worker Migration and Vulnerabilities to Trafficking 2010 [computer file]

Income expenditure trends

Respondents were asked what they spent the majority of their income on. The majority of both migrant and non-migrant respondents identified ‘supporting their family’ as their major expenditure, with significantly more migrant respondents than non-migrants indicating that they were supporting family overseas (Table 10).

Migrant respondents were significantly less likely to be spending the majority of their income on drugs (Table 10). This may reflect the extremely low rates of injecting and illicit substance use among Asian sex workers as measured in the SSHC survey (Pell et al. 2006).

Table 9 How respondents spend the majority of their income by migrant status (%)
Migranta Non-migrantb Significance testingc
Support myself and my family in Australia 43 76 Yates’ adj χ=35.4**
Save [saving money] 34 41 ns
Education fees 30 20 ns
Support family in home country or country other than Australia 36 3 Yates’ adj χ(1)=43.7**
Pay debts in Australia 7 32 Yates’ adj χ(1)=50.0**
Pay debts in home country 14 1 Yates’ adj χ(1)=13.6*
Gamble 8 5 ns
Buy drugs 0d 16 Yates’ adj χ(1)=55.4**
Other 6 11 ns

* p<0.001

** p<0.0001

ns: not significant

a: Percentages calculated from the total number of migrant respondents who answered the question (n=405), excluding 7 migrant respondents who did not respond to this question

b: Percentages calculated from the total number of non-migrant respondents who answered the question (n=109), excluding 42 non-migrant respondents who did not respond to this question

c: Migrant status comparisons should be interpreted with caution as non-migrants were significantly less likely than migrants to answer the question on income expenditure

d: n=1, rounded to 0 percent

Note: Respondents could select multiple responses for the question on income expenditure

Source: AIC, Sex Worker Migration and Vulnerabilities to Trafficking 2010 [computer file]

Non-migrants were significantly more likely to be spending the majority of their income on debts (either in Australia, overseas or both) compared with migrant respondents (33%, n=36 cf 20%, n=79; Yates adj χ2(1)=8.28, p<0.01). Respondents who indicated that they spent the majority of their income on debt in Australia or in their home country, or both (n=120, including 5 respondents with an unclassified migrant status), were also asked whether this debt was incurred by travelling to Australia or securing their current job. Thirty percent (n=31) of the 104 respondents who answered this question indicated that their debt was incurred by travelling to Australia or securing their current job. All but one of these respondents were migrants. Thus, seven percent of the entire migrant survey sample indicated that they spent the majority of their income on debt incurred by travelling to Australia or securing their current job.

The New Zealand survey asked migrant respondents the same question on income expenditure. The New Zealand responses reflected those of this survey, with supporting themselves and their family in New Zealand (43%) and saving money (28%) emerging as the major expenditures. Key differences between this survey and the New Zealand survey included less New Zealand migrant sex workers spending the majority of their income on supporting family members in another country (28% cf 36%) (see Table 9), paying debts in another country (7% cf 14%), education fees (19% cf 30%) and gambling (2% cf 8%).

The New Zealand survey (see Appendix D) also asked respondents why they stayed in the industry, with a range of multiple-choice options provided that related to income expenditures and workplace conditions. These responses were compared with the responses of non-migrant sex workers in New Zealand from a previous study (Roguski 2013). Reflecting the areas of major income expenditure, payment of household expenses (76% migrants, 82% non-migrants) and supporting children/family (49% migrants, 40% non-migrants) emerged as the common reasons for staying in the industry (Roguski 2013). Supporting alcohol or other drug use was less of a reason for working in the industry for migrant respondents (4%) than for non-migrants (17%; Roguski 2013), reflecting the differences found between income expenditures of migrants and non-migrants in this survey (Table 9).

Further differences between migrant and non-migrant respondents in the New Zealand studies include significantly less migrant respondents staying in the industry to support a gambling habit (1% cf 39%; Roguski 2013). This proportion of non-migrant respondents who stayed in the industry to support their gambling habit was substantially higher than the non-migrant respondents in this AIC survey who stated that gambling was a major income expenditure (5%; see Table 9). However, highlighting the conceptual differences between the question on reasons for staying in the industry and the question on major income expenditures, migrant respondents in the New Zealand study were more likely to state that paying for education was a reason for staying in the industry (40%) than that it was a major income expenditure (19%). Similarly, they were more likely to state that saving was a reason for staying in the industry (59%) than that saving was a major income expenditure (28%; Roguski 2013).