Grazyna S. Majchrzak-Hamilton and Neil T. M. Hamilton

This article presents a detailed examination of survey data on the circumstances of two groups of indigenous people: one consisting of indigenous people who, in their childhood, had been separated from their parents and one consisting of those who had not been separated. In almost all cases the separated group was worse off. On average they had left school earlier, were less likely to have educational qualifications and were less likely to be employed.

INTRODUCTION

Australia’s indigenous Stolen Generations (those children forcibly removed from their families by government action) have been one of the key areas of interest in the current debate surrounding social reconciliation. The culmination of this attention has been the Human Rights and Equal Opportunity Commission’s (HREOC) report entitled Bringing Them Home,¹ a comprehensive account of the history and qualitative effects of forcible indigenous child removal.

The HREOC inquiry has been of massive importance in achieving a greater understanding of the consequences to indigenous people of government policy and social attitudes. The Inquiry’s terms of reference, however, did not include any specific assessment of the effects of child removal policies on the indigenous population. It is therefore not surprising that only three and a half pages of the report are devoted to this important area, two of which contain tables from the National Aboriginal and Torres Strait Islander Survey (NATSIS), conducted by Australian Bureau of Statistics in 1994.² This ABS survey provided the first real opportunity for quantitative research on the effects of child removal policies.

The results presented in the tables in the HREOC report show:³

• ‘no significant difference between the two (taken-away and not-taken-away) groups with respect to their educational achievement’;
• that ‘the group removed from their families in childhood was no more likely to be employed ... than people who were not removed’;
• ‘Neither are people removed as children significantly more likely to earn higher incomes’.

• ‘those removed in childhood were twice as likely to have been arrested more than once in the last 5 years’; and
• ‘the same factors have an effect on health prospects’.

The key question from the NATSIS which underpins the data analysis presented here is:

The next question is about forced separation of children from their families. Please tell me if you don’t want to answer this question.

Were you taken away from your natural family by a mission, the Government, or welfare?

DATA

Data source

The National Aboriginal and Torres Strait Islander Survey was conducted by the Australian Bureau of Statistics (ABS) in 1994, following the 1987 Royal Commission into Aboriginal Deaths in Custody. Preliminary results were released at the beginning of 1995.⁴ The current study became possible in 1996 after the release of the confidentialised unit record file (CURF) for use by the wider public, and this is the first major publication relating to the Stolen Generation using these data.

The inclusion of data on child separation in the NATSIS was seen as a first national indicator of the extent of forced separation of children from their families.⁵ Previous estimates were made on the basis of historical evidence and reports from institutions involved in implementing the policy of child removal.

The ABS employed a multistage stratified sample design, with stratification based on Aboriginal and Torres Strait Islander Commission (ATSIC) regions, section of state, and Aboriginal population in each Census Collector District (CD) (on the basis of the 1991 Population Census). Enumeration of all Aboriginal and Torres Strait Islander people within selected households was undertaken. The sample size was 6.6 per cent of the ATSI population as reported in the 1991 census, or just over 18,000 individuals. Of this number, people of non-ATSI origin or less than 12 years old were not asked if they had been separated from their families, leaving a sample of 15,527 people. Approximately two per cent of this number declined to answer the question about separation. Most of the results presented in this paper refer to the indigenous population over the age of 20 who answered the questions regarding separation. This sub-group comprised 7,380 persons in the NATSIS.

Data limitations

Data from NATSIS, as with any data from a sample survey, are prone to sampling errors. Despite the sampling strategy being selected to meet time and budget constraints, the data at a national level are perceived to be of high quality.⁶ Errors of under-enumeration of separated people were most likely to occur in CDs with low numbers of indigenous people.

Non-sampling errors are also of significant concern in the study. The changing role of self-identification with time is a known and difficult issue. Misunderstanding of the crucial question about separation is likely to have occurred through different interpretations, lack of knowledge or recollection, or general confusion. These, and other, factors contribute to a higher than average level of errors in some portions of the NATSIS results.

FINDINGS

Separation rates
 

Table 1: The estimated ATSI population, per cent of persons who had been separated from their natural families by age and sex groups, 1994
Age (years)	Male	Female	Total		Sex ratios
	% Sep.	% Sep.	% Sep.	Estimated Population	Sep.	Total
0-12 13-19 20-34 35-49 50+	1.2 3.3 8.5 10.5 11.1	1.6 3.6 7.2 11.6 10.9	1.4 3.4 7.9 11.0 11.0	105,718 43,804 83,063 44,296 26,369	80.4 95.6 120.7 86.5 86.8	105.9 103.6 102.2 95.4 85.4
Total	5.6	5.8	5.7	303,250	98.4	101.0
Source: Computed from National Aboriginal and Torres Strait Islander Survey (NATSIS) 1994 Estimated population refers to the ABS projection of the total indigenous population for 1994 which was used as a basis for NATSIS. Total sample size was just over 18,000, of whom 15,527 identified as Aboriginal or Torres Strait Islander (ATSI) over the age of 12. Sex ratios: mean number of males per 100 females Sep. stands for persons separated from their natural families

Separation of children from natural families is not an issue from the past. NATSIS data indicate that, although declining, the act of separation was reported by all age groups in 1994. In the group of children 0-12 years old, 1.4 per cent reported as being separated from their natural parents (Table 1). This proportion increases with age, reaching 11 per cent among persons aged 35 years or older. Overall, 5.7 per cent (estimated to represent more than 17,000 persons) of the indigenous population reported separation from their natural families. This figure has to be viewed as the lower boundary of estimates due to the survey technique, as it is based on self-reported data of survivors only.

It is expected that a significant proportion of people affected by separation are no longer alive due to abnormally high levels of adult mortality within the indigenous population.⁷ Further, more than one per cent of the total indigenous population declined to discuss the matter during an interview. It is estimated that, at certain times, approximately one in three indigenous children were removed from their natural families.⁸ This can easily be equated to approximately 100,000 indigenous people affected, half of whom may be still unaware of their origins or choose not to self-identify as Aboriginal people.⁹ The HREOC inquiry¹⁰ concluded that between 10 per cent and 30 per cent of all Aboriginal and Torres Strait Islander people were forcibly removed from their families between 1910 and 1970, and that not a single family had escaped the effects of forcible removal.
 

Figure 1: Spatial distribution of Aboriginal and Torres Strait Islander population by part of state, gender and whether separated or not

The spatial distribution of the indigenous population throughout Australia differs markedly from the distribution of the total Australian population,¹¹ the majority of the indigenous population tending to live outside the state capital cities. This pattern is reflected in the total population who experienced separation from their natural families; however, as Figure 1 illustrates, a higher proportion of people who were separated from their natural families live in capital cities than those who did not experience separation. The largest proportion of people separated from their natural family in each age group live in ‘other urban’ areas, similar to the population which did not experience separation. This pattern differs by gender and age.¹²

Education
 

Table 2: The estimated ATSI population by whether separated or not, age, and age left school, 1994 (percentages)
Age left school	Age (years)
	20-34		35-49		50+		Total
	Sep.	Not	Sep.	Not	Sep.	Not	Sep.	Not
Under 16 years 16 years or more Never went to school Still at school Not stated	46.0 49.5 0.6 2.7 1.4	36.0 62.6 1.1 0.1 0.2	62.0 34.5 1.5 0.0 2.0	58.4 36.3 4.6 0.0 0.6	62.1 12.9 23.3 0.0 1.7	64.2 13.2 21.8 0.0 0.7	54.6 36.9 5.5 1.2 1.7	47.1 46.9 5.6 0.1 0.4
Total per cent Total number	100.0 6,534	100.0 74,897	100.0 4,894	100.0 38,743	100.0 2,888	100.0 22,927	100.0 14,316	100.0 136,567
Source: See Table 1, but sample size in Table 2 is 7,380 Notes: Columns totals may not add up to 100 per cent due to rounding

At the 1991 census, over 40 per cent of indigenous people aged 15 years or more reported leaving school before the age of 16, and 4.6 per cent had never attended school, a longstanding and well-documented problem.¹³ This pattern is reflected in the current study. More than half (54.6 per cent) of the group aged 20 years or over who were separated from their natural families reported leaving school before the age of 16 (Table 2). This compares with 47.1 per cent of those who were not separated. A higher proportion of people separated from their families left school earlier than those who were not separated, except in the cohort aged 50 years or more. A higher proportion of those not taken away stayed at school until they were 16 years or older compared to those who were taken away.

More than five per cent of both groups reported never attending school, the proportion increasing with age in both groups (Table 2). This proportion is much lower for the younger cohorts of the groups who have been separated from their families than for the same cohorts in the group who were not separated. The trend reverses for the oldest cohort, where 23.3 per cent of the people who experienced separation indicated never having attended school, compared with 21.8 per cent of the group who have not been separated.

The majority of the separated group obtained less than 10 years of schooling (Table 3). There is a distinct pattern by age group of higher proportions of people who were separated obtaining less than 10 years of schooling compared with the people who were not. The year-12 certificate was also obtained by a lower proportion of the group who had experienced forced separation compared with the population who had not. An exception is the 35-49 age group, where 6.6 per cent of the people who experienced forced separation completed the year-12 certificate compared with 3.6 per cent of the people who did not experience separation.
 

Table 3: The estimated ATSI population by whether separated or not, age and educational attainment, 1994 (percentages)
Age left school	Age (years)
	20-34		35-49		50+		Total
	Sep.	Not	Sep.	Not	Sep.	Not	Sep.	Not
School attainment: Year 12 school certificate Year 10 school certificate Schooling below Year 10	6.8 49.9 39.7	14.7 47.9 35.6	6.6 25.9 64.0	3.6 30.0 60.7	0.0 3.4 71.0	0.8 7.8 67.0	5.4 32.3 54.3	9.2 36.1 48.0
Post school qualifications	18.1	20.4	18.5	20.4	7.7	8.4	16.1	18.4
Total per cent Total number	100.0 6,534	100.0 74,897	100.0 4,894	100.0 38,743	100.0 2,888	100.0 22,927	100.0 14,316	100.0 136,567
Source: As for Table 2. Notes: Underlined figures are statistically significant at p< .01.

At the time of the 1991 Census, more than 79 per cent of indigenous people aged 15 years and over reported being not qualified.¹⁴ In general, a low percentage of indigenous people attend universities¹⁵ or obtain post-school qualifications.¹⁶ In the current study in all age categories, lower proportions of people who were separated from their natural families held post-school qualifications. However, the qualifications reported by the separated group were of a higher level (falling into the categories of undergraduate, diploma and skilled vocational) than those reported by people not separated, which lay predominantly in the categories of unskilled vocational and inadequately described.

Labour-force status

Labour-force participation and employment

In 1994, only 31.3 per cent of people who had been separated from their natural families reported that they were employed, compared with 38.8 percent of those who had not been not separated (Table 4). This includes both Community Development Employment Projects (CDEP) and non-CDEP jobs. The proportion of people in CDEP employment was 8.5 per cent of those who were separated from their natural families and 9.1 per cent for those who were not. The difference between the two groups is even greater when we look at people in non-CDEP employment. Here a much lower proportion of people who had experienced separation were employed in non-CDEP jobs (22.8 per cent) compared with those who had not (29.7 per cent). This pattern persists for all age categories for non-CDEP employment.
 

Table 4: The estimated ATSI population by whether separated or not, age and labour force status,1994 (percentages)
Age left school	Age (years)
	20-34		35-49		50+		Total
	Sep.	Not	Sep.	Not	Sep.	Not	Sep.	Not
Employed Employed - CDEP Employed - non CDEP Employed subtotal	9.1 21.3 30.4	10.6 29.3 39.9	9.9 31.2 41.1	8.7 38.4 47.1	4.8 12.1 16.9	4.6 16.2 20.8	8.5 22.8 31.3	9.1 29.7 38.8
Unemployed	29.8	27.8	23.4	17.6	3.0	5.4	22.2	21.1
Not in the labour force	29.5	29.9	34.2	34.8	79.5	73.6	41.2	38.6
Not applicable	10.2	2.4	1.3	0.5	0.7	0.1	5.2	1.5
Total per cent Total number	100.0 6,534	100.0 74,897	100.0 4,894	100.0 38,743	100.0 2,888	100.0 22,927	100.0 14,316	100.0 136,567
Source and notes: As for Tables 2 and 3.

Unemployment

Analysis of NATSIS data shows that the proportion of people who were unemployed at the time of the survey was similar for both the people who had been separated from their natural families and those who had not (22.2 per cent and 21.1 per cent respectively). However, it is notable that a higher proportion of the younger cohorts of people who had been separated from their natural families were unemployed (Table 4). This finding is crucial to the assessment of past government policies of assimilation involving forced child removal and placement in the custody of the State.

Income

The median income calculated for both the separated and non-separated groups in this study was just over $11,000, and is similar for each group when examined by age group (Table 5).

Disaggregation of the income figures, however, reveals significant differences between the target groups. For example, the median income from non-CDEP employment differs markedly between them. Overall, the non-CDEP median income of the group separated from their natural families was higher, just above $26,000, compared with $23,350 for those not separated. Although there was a lower proportion of people who had experienced forced separation employed in non-CDEP jobs, the median income for those people in each age group was higher than for those not separated.

CONCLUSIONS
 

Table 5: The estimated ATSI population by whether separated or not, age, income and main source of income, 1994 (medians and percentages)
Age left school	Age (years)
	20-34		35-49		50+		Total
	Sep.	Not	Sep.	Not	Sep.	Not	Sep.	Not
Median income Personal income CDEP Non-CDEP	$11,520 $9,170 $29,320	$11,570 $10,170 $23,210	$13,090 $15,270 $27,270	$13,280 $11,440 $23,510	$9,430 $9,540 $23,510	$9,110 $11,400 $24,200	$11,180 $10,790 $26,010	$11,220 $10,580 $23,350
Main source of income: % Wages and salaries Business income Government payments No income Not stated	28.0 1.0 56.8 12.7 1.5	35.2 1.8 54.0 7.6 1.4	34.0 0.6 56.6 2.8 6.0	39.0 2.9 49.4 6.9 1.8	14.5 1.0 79.0 4.5 1.0	17.8 1.0 74.1 5.6 1.5	27.3 0.9 61.2 7.7 3.0	33.4 2.0 56.1 7.1 1.5
Total per cent	100.0	100.0	100.0	100.0	100.0	100.0	100.0	100.0
Total number	6,534	74,897	4,894	38,743	2,888	22,927	14,316	136,567
Source and notes: As for Tables 2 and 3.

The taking away of indigenous people from their natural families is not only affecting past generations, it is still proceeding. The trend has been declining and the types of institutions involved in raising these people have changed over time. The current study of the NATSIS data reveals that there is little evidence of any positive outcome of Australia’s child removal policies. In general, these policies did not enhance indigenous people’s level of education or employment prospects. The level of education obtained by the group separated from their natural families is lower than the level achieved by those who were not separated. This poses the question of the quality and duration of education provided by various institutions in the past. The lack of positive outcomes in the employment area is also important. The low proportion of people separated from their natural families employed in mainstream jobs (non-CDEP) and the high proportion relying on Government benefits is a testimony to the total failure of child removal policies.

The picture presented here, then, is quite different from the general NATSIS results presented in Bringing Them Home,¹⁷ which clearly focussed on providing an impression of no socio-economic deprivation for the Stolen Generation. The present study illustrates the importance of a thorough understanding and analysis of the data, and the dangers of presenting highly summarised accounts of complex issues. Of greater relevance, perhaps, is the fact that there are now data available to investigate socio-economic factors and their interrelationship with indigenous childhood separation and to use both to shed light on outcomes in adult life. Such analysis clearly shows differing results for the two groups, and in almost every case the Stolen Generation have clearly suffered from their experiences.

Acknowledgments

This study was undertaken by one of the authors (G.S.M.-H.) as part of a Masters degree in Demography in the Research School of Social Sciences at ANU, supervised by Dr Habte Tesfagiorghis and advised by Dr Peter Read. An earlier draft of the paper has been improved by comments from Dr Doug Cocks, and two anonymous referees.

References  

^{1 Human Rights and Equal Opportunity Commission (HREOC), Bringing Them Home: Report of the National Inquiry into the Separation of Aboriginal and Torres Strait Islander Children from their Families, HREOC, Sydney, 1997}

^{2 Australian Bureau of Statistics (ABS), National Aboriginal and Torres Strait Islander Survey 1994, Detailed Findings, Canberra, 1995; J. Altman and J. Taylor, The 1994 National Aboriginal and Torres Strait Islander Survey: Findings and Future Prospects, Centre for Aboriginal Economic Policy Research, Australian National University (ANU), 1996 3 HREOC, op. cit, pp. 14-15}

^{4 ABS, 1995, op. cit.}

^{5 ABS, National Aboriginal and Torres Strait Islander Survey, An Evaluation of the Survey. Canberra, 1996, p. 50}

^{6 ibid.}

^{7 Australian Institute of Health and Welfare (AIHW), Australia’s Health 1992, Australian Government Publishing Service (AGPS), Canberra, 1992}

^{8 P. Read, 1996, personal communication}

^{9 ibid.}

^{10 HREOC, op. cit.}

^{11 ABS, Australia in Profile, 1991 Census. Canberra, 1993}

^{12 G. Majchrzak-Hamilton, Australia’s Stolen Generations: the legacy of previous governments’ policies for the indigenous population of Australia, unpublished Master of Arts thesis, Research School of Social Sciences, ANU, Canberra, 1997}

^{13 A. E. Daly and L. Jin, Estimating the private rate of return to education for indigenous Australians, Discussion Paper No. 97, Centre for Aboriginal Economic Policy Research, ANU, Canberra, 1995}

^{14 ABS, Australia’s Aboriginal and Torres Strait Islander Population, 1991 Census, Canberra, 1993}

^{15 A. E. Daly, Education and employment for young indigenous Australians, 1986 to 1991. Discussion Paper No. 50, Centre for Aboriginal Economic Policy Research, ANU, Canberra, 1993}

^{16 ABS, 1995, op.cit.}

^{17 HREOC, op. cit.}
 

Back to People and Place Home Page