bs_bs_banner
Journal of Intellectual Disability Research 530
doi: 10.1111/jir.12158
volume 59 part 6 pp 530–540 june 2015
Ethnicity and self-reported experiences of stigma in adults with intellectual disability in Cape Town, South Africa A. Ali,1 E. Kock,2 C. Molteno,3 N. Mfiki,4 M. King1 & A. Strydom1 1 Division of Psychiatry, University College London, London, UK 2 Western Cape Forum for Intellectual Disability, Alexandra Hospital, Maitland, South Africa 3 Department of Psychiatry and Mental Health, University of Cape Town, Groote Schuur Hospital, Observatory, South Africa 4 Alexandra Hospital, Maitland, South Africa
Abstract Background Studies have shown that individuals with intellectual disability (ID) are aware of stigma and are able to describe experiences of being treated negatively. However, there have been no crosscultural studies examining whether self-reported experiences of stigma vary between ethnic groups. Method Participants with mild and moderate ID were recruited from a number of different settings in Cape Town, South Africa. Self-reported experiences of stigma in three ethnic groups were measured using the South African version of the Perceived Stigma of Intellectual Disability tool, developed by the authors. One-way anova was used to test whether there were differences in the total stigma score between the ethnic groups. Regression analysis was performed to identify factors associated with stigma. Results A total of 191 participants agreed to take part; 53 were Black, 70 were of mixed ethnicity and 68 were Caucasian. There were no differences in the levels of stigma reported by the three groups but the Black African ethnic group were more likely to report being physically attacked and being stared at, Correspondence: Dr Afia Ali, Division of Psychiatry, University College London, 2nd Floor, Charles Bell House, 67-73 Riding House Street, London W1W 7EY, UK (e-mail: [email protected]).
but were also more likely to report that they thought they were ‘the same as other people’. There was an interaction effect between ethnicity and level of ID, with participants with mild ID from the Black African group reporting higher levels of stigma compared with those with moderate ID. Younger age was the only factor that was associated with stigma but there was a trend towards ethnicity, additional disability and socio-economic status being related to stigma. Conclusion Interventions should target the Black African community in South Africa and should include the reduction of both public stigma and self-reported stigma. Keywords discrimination, ethnicity, intellectual disability, negative attitudes, stigma
Introduction Conceptualising stigma People with intellectual disability (ID) are one of the most stigmatised and excluded groups in society (European Union Monitoring and Advocacy Program 2005) and are more likely to be subjected to physical violence compared the general population (Hughes et al. 2012). The conceptualisation of
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research
volume 59 part 6 june 2015
531 A. Ali et al. • Ethnicity and stigma in intellectual disability
public or social stigma has evolved over the last five decades. Goffman’s (1963) characterisation of stigma as an attribute that is deeply discredited by a person’s society, resulting in his/her rejection and devaluation, has been argued by many researchers to have limited utility in understanding healthrelated stigma because of its outdated use of language, generalised application to a wide range of situations and focus on social interactions rather than political or structural forces (Weiss et al. 2006). Stigma is now considered to be influenced by structural and cultural forces. Link & Phelan (2001) went on to describe stigma as a process defined by labelling, stereotyping, separating, status loss and discrimination, occurring in the presence of an imbalance of political and economic power. The social cognitive model of stigma, formulated by Corrigan and colleagues, suggests that stigma comprises of cognitive, emotional and behavioural aspects, manifesting as stereotypes, prejudice and discrimination (Corrigan 2000; Corrigan & Watson 2002). At the individual level, stigma may become internalised (also known as self or felt stigma), which occurs when the individual endorses cultural stereotypes about their condition, applies it to him/ herself and believes that they will be devalued by others (Corrigan et al. 2006).
analysis of results from a large population-based study of 1273 participants with ID (aged over 16) residing in at home. Questions were asked about whether they had experienced bullying at school and whether people had been rude or nasty to them in the last 12 months. Half the sample reported that they had been bullied at school and a third reported that they had been the recipient of bullying in the last 12 months. Participants with ID have also reported discrimination in finding employment and from colleagues at work because of misconceptions about abilities (Li 2004). Jahoda & Markova (2004) found that participants who had recently moved from institutions into community settings, deliberately concealed their past in order to be accepted. Living independently was perceived as a positive step in overcoming stigma. The experience of stigma, particularly if it has been internalised, has been linked to poor psychological well-being in people with ID. Higher levels of reported stigma is associated with lower selfesteem (Szivos-Bach 1993; Abraham et al. 2002; Paterson et al. 2012), comparing oneself less favourable to others in the community (Dagnan & Waring 2004; Cooney et al. 2006; Paterson et al. 2012) and psychological distress (Emerson 2010; Paterson et al. 2012).
The experience of stigma by people with ID
Public’s knowledge and attitudes towards ID
Studies show that people with ID are aware of stigma and that they are viewed negatively by society. Cooney et al. (2006) examined whether there were differences in the levels of stigma experienced by students attending segregated schools and mainstream schools. The students attending mainstream schools were more likely to report stigmatising treatment compared with those attending segregated schools. However, both groups reported similar experiences outside the school environment. Chen & Shu (2012) found that students with ID attending a mainstream school in Taiwan reported feelings of shame and embarrassment as a result of ‘odd stares’ and ‘strange looks’ looks from their peers. The high value placed on academic ability in Taiwan and similar cultures may result in people with ID being regarded more negatively compared with other cultures that place less emphasis on education. Emerson (2010) conducted a secondary
The experience of stigma and discrimination by marginalised groups is closely related to public stigma, for example higher levels of stigma are reported by people with mental illness in countries with higher levels of public stigma towards mental illness (Vogel et al. 2013). It is therefore helpful to examine the literature in relation to public stigma towards people with ID. Scior (2011) conducted a systematic review, which found that negative attitudes towards people with ID were apparent across different cultures. Several studies consistently demonstrate that members of the lay public have a limited understanding of the concept of ID (Gordon et al. 2004), and that awareness and knowledge varies between respondents from different cultural and ethnic backgrounds. Knowledge about the prevalence of ID is often inaccurate (Alem et al. 1999; Tachibana & Watanabe 2003; Tachibana 2006). One study found that only one in
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 6 june 2015
Journal of Intellectual Disability Research 532 A. Ali et al. • Ethnicity and stigma in intellectual disability
four people were able to recognise a person with ID from a vignette. Recognition was highest among White British participants compared with those from Asian and African backgrounds (Scior & Furnham 2011). There are misconceptions about the aetiology of ID. In India, common explanations are that it was ‘God’s will’ or the fault of the parents (Madhavan et al. 1990) and in Tanzania, witchcraft was commonly cited as a cause by tribal leaders (Kisanji 1995). Studies examining social distance suggest that the general public are unlikely to consider someone with ID as a friend (Gordon et al. 2004). Attitudes towards people with more severe ID tend to be more negative compared with those with mild ID (Antonak et al. 1995) and are linked to perceptions about the ability and competence of people with ID. Attitudes in developed countries generally promote the inclusion of people with ID in society (Bryant et al. 2006; Henry et al. 1996). Negative attitudes are more likely to be associated with supernatural forces (Mulatu 1999) and views about the condition being self-inflicted (Panek & Jungers 2008). A few studies of cross-cultural comparisons suggest that some cultures may hold particularly negative attitudes. For example, Asian American students held more negative attitudes than Latin American or African students (Saetermore et al. 2001) and White British respondents were more positive towards people with ID compared with respondents from Hong Kong (Scior et al. 2010).The most consistent factor associated with positive attitudes is having prior contact with someone with ID. In particular, positive attitudes are associated with having positive contact (Morin et al. 2013).
Variables associated stigma and double stigma Studies of people with ID have not found a relationship between stigma and age (Szivos-Bach 1993; Cooney et al. 2006; Paterson et al. 2012), stigma and gender (Szivos-Bach 1993; Petrovski & Gleeson 1997; Dagnan & Waring 2004; Cooney et al. 2006; Paterson et al. 2012) or between stigma and IQ (Szivos-Bach 1993; Cooney et al. 2006; Paterson et al. 2012) possibly because the studies were not sufficiently powered to detect differences (sample size of the studies were 50 participants or less).
There is little research on the relationship between ethnicity, culture and the experience of stigma experienced by people with ID. In other areas of research, cultural factors have been found to influence the experience of stigma. For example one study of older adults found that Asian Americans and Latinos expressed greater shame and embarrassment about having mental illness compared with non-Latino Whites (Jimenez et al. 2013). Depressed women of immigrant African and Caribbean backgrounds, and USA-born Black or Latino women, were less likely to seek treatment compared with USA-born White women because of the fear of stigma (Nadeem et al. 2007). People with ID may have other potentially stigmatising features such as belonging to a non-White ethnic minority group, the presence of mental illness or behavioural problems, mobility and speech problems. The presence and combined effects of more than one stigmatising condition is known as ‘double stigma’, and may lead to individuals reporting higher levels of stigma compared with those who have just one stigmatising condition. There is relatively little research on the impact of multiple stigmatising characteristics such as the impact of ethnicity or racism and ID on the experience of stigma. However, one qualitative study of South Asians with ID living in the UK, suggests that the combined effects of racism and the presence of ID contributed to higher perceptions of discrimination and harassment (Azmi et al. 1997). In other areas of research, a higher perception of racial discrimination among Black women in Canada, has been correlated with higher HIV stigma (Logie et al. 2013).
Rationale for study Given that there is evidence for cultural differences in attitudes towards people with ID, one might expect there to be cultural differences in the experience of stigma by people with ID. However, there is little research in this area and no cross comparisons have been made examining the experience of stigma in people with ID from different communities. As there is relatively little known about ethnic or cultural differences in the experience of stigma, our study was aimed at examining differences in the reported experiences of stigma in people with ID from three ethnic groups in Cape
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 6 june 2015
Journal of Intellectual Disability Research 533 A. Ali et al. • Ethnicity and stigma in intellectual disability
Town, South Africa. Given the historical context of Apartheid South Africa, this has led to ethnic groups developing distinct cultural identities, which makes this area a unique location to study differences in stigmatisation.
Objectives 1 To identify whether there are differences in levels of stigma reported by three main ethnic groups in Cape Town: Xhosa-speaking Black Africans; those of mixed ancestry; and Afrikaans or Englishspeaking Caucasians of White European Descent; 2 To identify whether there are differences in selfreported stigma between those with mild or moderate ID, and between participants from other potentially stigmatising groups (e.g. those with physical disability, an unusual appearance, mental illness or behavioural problems); and 3 The assess for interaction effects between ethnicity and level of ID on self-reported stigma.
Hypotheses 1 Levels of self-reported stigma will differ between the three ethnic groups. The Xhosa-speaking Black African group will report higher levels of stigma because of the combined effects of potential racial discrimination and ID. 2 Levels of reported stigma will be higher in those with moderate ID compared with mild ID because the moderate group may have more ‘noticeable disability’. 3 Levels of self-reported stigma will also be higher in other groups with potentially stigmatising attributes. 4 The effect of ethnicity on self-reported stigma will be modified by level of ID.
Method Ethical approval Ethical approval was obtained from the Research Ethics Committee, Health Science Faculty, University of Cape Town. The study was conducted in the Western Cape Metropolitan Area, South Africa between August 2007 and May 2008.
Study area The study area was the metropolitan area of Cape Town in the Western Cape Province, South Africa. Cape Town has 3.4 million inhabitants, and includes the city of Cape Town, Mitchells Plain and Khayelitsha. During the 2007 community survey, 44% of the population indicated their population group as mixed ancestry, 35% was Black African, 19% was Caucasian and 2% were Asian. Afrikaans was the most common home language (41% of households) followed by Xhosa (29%) and English (28%) (Statistics South Africa 2009). The age distribution reflected a young population with 27% of people aged 14 years or younger. Sixty-five per cent of the population was younger than 35 years of age. In 2007 3.5% of the total population of Cape Town had a physical and mental disability.
Participants Potential participants with mild or moderate ID were identified by occupational therapists, supervisors, nurses or social workers attached to various sheltered employment workshops, outpatient clinics, the provincial government or local voluntary organisations. Participants were recruited based on language as well as ethnic group to ensure that we had approximately equal numbers of participants per language and ethnic group. Background information was gathered from the participant, as well as their caregiver or other informant (e.g. the occupational therapist at the sheltered employment, or the psychiatric nurse at the outpatient clinic). Participants were interviewed at their workshops, homes or clinics. Adults with mild to moderate ID were included. Participants under the age of 18 were excluded from this study as the stigma measure (see below) had not been validated in this group, and those who were unable to provide informed consent were also excluded. One hundred and ninety-one participants were included in the study. Fifty-three (27.7%) were Black Africans; 68 (35.6%) were Caucasians; and 70 (36.6%) were of mixed ancestry. Women made up 55.5% of the sample. One hundred and six (55.5%) had a mild ID and the remainder had a moderate ID. Seventy-one (37%) spoke Afrikaans, 67 (34.9%) English and 51 (27.6%) Xhosa.
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 6 june 2015
Journal of Intellectual Disability Research 534 A. Ali et al. • Ethnicity and stigma in intellectual disability
Measures Self-reported experiences of stigma Self-reported experiences of stigma were measured with the South African version of the Perceived Stigma of Intellectual Disability tool (Kock et al. 2012) that was originally developed in the UK (Ali et al. 2008). The UK version was translated into English, Afrikaans and Xhosa by a team of experts who worked with adults with ID. Independent translators back translated the measure into English to ensure that the items retained their original meanings. The face validity of the stigma measure was assessed by participants with ID in three focus groups representing the three languages. The measure was then piloted by two participants from each language (see Kock et al. 2012 for further information about the development of the measure). The South African stigma measure is a self-report instrument with two sub-scales: ‘felt stigma’ and ‘reaction to felt stigma’ and each item has an accompanying picture, with a yes/no response format. Items from the ‘felt stigma sub-scale include: ‘I think I am the same as other people’, ‘people on the street make fun of me’, ‘people make fun of me for going to the day centre’. Items from the ‘reaction to felt stigma sub-scale’ include: ‘the way people talk to me makes me angry’ and ‘people make me feel embarrassed’. In order to ensure that the responses given were valid, the interviewer asked the participant to give examples of situations when the item under question occurred. The items are scored 0–10. The tool has good psychometric properties (per item test re-test reliability kappa 0.41 to 0.59; factor loadings 0.51–0.83; Cronbach’s alpha 0.73).
behaviour drew attention from their community. In addition, socio-economic status was recorded, which was defined by area of residence based on data obtained from the 2001 census (Statistics South Africa 2003; Gie & Ramonovsky 2006). We rated socio-economic status on a five-point scale (1–5) with a score of one indicating higher socioeconomic status.
Analysis Analysis of data was performed using SPSS, version 17. In order to simplify the analysis, the variables relating to stigmatising features (apart from level of ID) were combined into two composite variables: ‘obvious additional disability’, and ‘behaviour or mental disorder’ and the responses were dichotomised to ‘yes’ and ‘no’. The demographic characteristics of the whole sample and the three ethnic groups were analysed using descriptive statistics. Individual stigma items were analysed to assess whether there were differences in the responses given by the different ethnic groups. As the total stigma score was normally distributed, one-way analysis of variance (anova) was used to test whether there were statistically significant differences in the total stigma score between the three ethnic groups. Interaction effects between ethnic group and the demographic and clinical variables were assessed using two-way between-group analysis of variance. A multiple linear regression analysis was performed to identify factors associated with stigma and to adjust for variables that were found to confound the relationship between ethnicity and the total stigma score.
Results Socio-demographic and clinical variables
Sample characteristics
A structured data collection form was used to record information on gender, age, ethnicity, language (Afrikaans, English and Xhosa), level of ID (which was based on a clinical assessment using International Classification of Disease (ICD)-10 criteria (World Health Organisation 1993), sensory problems, mobility problems, involuntary movements, physical illness, mental illness (e.g. epilepsy), medication, whether the individual looked noticeably different from other people and whether their
Demographic factors, according to ethnic group are shown in Table 1. The Black African group had a significantly higher proportion of people with mild ID (Chi 22.9, P < 0.001) and almost all the participants were living in areas with the lowest socio-economic ratings, compared with the Caucasian group, where the majority of the participants were living in areas with the highest socio-economic ratings (Chi 296.1, P < 0.001). In addition the majority of the Black African group lived within the
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 6 june 2015
Journal of Intellectual Disability Research 535 A. Ali et al. • Ethnicity and stigma in intellectual disability
Total number Gender: males Age (mean) Severity of ID Mild ID Moderate ID Language Xhosa Afrikaans English Socio-economic group 1 2 3 4 5 Accommodation Own home Family home Group home Foster care Mobility problems Unusual appearance Speech problems Behaviour or mental disorder
Black African Number (%) 53
Mixed race Number (%) 70
Caucasian Number (%) 68
24 (45.3) 25.7
30 (42.9) 32.7
31 (45.6) 34.7
40 (75.5) 13 (24.5)
39 (55.7) 29 (41.4)
27 (39.7) 31 (45.6)
53 (100) 0 (0) 0
0 (0) 52 (74.3) 18 (25.7)
0 (0) 19 (27.9) 49 (72.1)
0 0 0 1 (1.9) 52 (98.1)
3 (4.3) 6 (8.6) 24 (34.3) 37 (52.9) 0
38 (56.7) 22 (32.8) 2 (3.0) 4 (6.0) 1 (1.5)
0 52 (98.1) 1 (1.9) 0 2 (3.8) 8 (15.1) 2 (3.8) 12 (22.6)
44 (62.9) 25 (35.7) 1 (1.4) 14 (20.0) 22 (31.4) 19 (27.1) 33 (47.1)
1 (1.5) 34 (50.0) 33 (48.5) 0 7 (10.3) 27 (39.7) 14 (20.6) 41 (60.3)
family home, compared with other groups (Chi 35.92, P < 0.001). In general, the Black African group had fewer additional stigmatising features (e.g. mobility problems, speech and behavioural problems, unusual appearance, mental illness and epilepsy).
Distribution of individual stigma items and total scores across the three groups Only three items were scored significantly differently between the three ethnic groups (see Table 2). The black African group were more likely than the other two ethnic groups to report that they had been attacked by people on the street and that they had been stared at. However, they were also more likely to report that they thought they were the same as other people. The total stigma scores did not differ between the three groups (F = 0.62, P = 0.54). Analysis of interaction effects on level of stigma revealed that there was a significant interaction
Table 1 Socio-demographic characteristics of the three ethnic groups
between ethnic group and level of ID (F = 3.74, P = 0.01). A graphical plot (Fig. 1) revealed that participants with mild ID from the Black African group reported higher levels of stigma than those with moderate ID and this effect was less marked for the Caucasian group. Those with moderate ID from the mixed ethnic group were more likely to report higher levels of stigma than those with mild ID, although the difference in the scores is not as striking as the Black ethnic group.
Factors associated with stigma Table 3 shows the results of the regression analysis, with total stigma score as the dependent variable. The independent variables included were age, gender, ethnicity, level of ID, socio-economic status, the presence of a behaviour or mental disorder, and obvious additional disability. Lower age was significantly associated with stigma (P = 0.02); however, there was a strong trend towards an
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 6 june 2015
Journal of Intellectual Disability Research 536 A. Ali et al. • Ethnicity and stigma in intellectual disability
Table 2 Responses to individual items on the stigma instrument given by the three ethnic groups
Item
Black Number (%)
Mixed Number (%)
White Number (%)
Chi square (P value)
I think I am the same as other people The way people talk to me makes me angry People make me feel embarrassed People on the street make fun of me People on the street have hurt me People on the street look at me in a funny way People treat me like a child People laugh at me because of the way I talk I worry about the way people act towards me People make fun of me about going to the day centre
47 (88.7) 34 (64.2) 31 (58.5) 25 (47.2) 24 (45.3) 33 (62.3) 24 (46.2) 24 (46.2) 25 (48.1) 18 (34.0)
35 (50.0) 48 (68.6) 38 (54.3) 34 (54.3) 8 (11.4) 32 (45.7) 26 (37.1) 26 (37.1) 28 (40.0) 22 (31.4)
34 (50.0) 49 (72.1) 57.4 (39) 24 (35.3) 8 (11.8) 27 (39.7) 25 (36.8) 25 (36.8) 25 (36.8) 17 (25.0)
24.02 (
Journal of Intellectual Disability Research 530
doi: 10.1111/jir.12158
volume 59 part 6 pp 530–540 june 2015
Ethnicity and self-reported experiences of stigma in adults with intellectual disability in Cape Town, South Africa A. Ali,1 E. Kock,2 C. Molteno,3 N. Mfiki,4 M. King1 & A. Strydom1 1 Division of Psychiatry, University College London, London, UK 2 Western Cape Forum for Intellectual Disability, Alexandra Hospital, Maitland, South Africa 3 Department of Psychiatry and Mental Health, University of Cape Town, Groote Schuur Hospital, Observatory, South Africa 4 Alexandra Hospital, Maitland, South Africa
Abstract Background Studies have shown that individuals with intellectual disability (ID) are aware of stigma and are able to describe experiences of being treated negatively. However, there have been no crosscultural studies examining whether self-reported experiences of stigma vary between ethnic groups. Method Participants with mild and moderate ID were recruited from a number of different settings in Cape Town, South Africa. Self-reported experiences of stigma in three ethnic groups were measured using the South African version of the Perceived Stigma of Intellectual Disability tool, developed by the authors. One-way anova was used to test whether there were differences in the total stigma score between the ethnic groups. Regression analysis was performed to identify factors associated with stigma. Results A total of 191 participants agreed to take part; 53 were Black, 70 were of mixed ethnicity and 68 were Caucasian. There were no differences in the levels of stigma reported by the three groups but the Black African ethnic group were more likely to report being physically attacked and being stared at, Correspondence: Dr Afia Ali, Division of Psychiatry, University College London, 2nd Floor, Charles Bell House, 67-73 Riding House Street, London W1W 7EY, UK (e-mail: [email protected]).
but were also more likely to report that they thought they were ‘the same as other people’. There was an interaction effect between ethnicity and level of ID, with participants with mild ID from the Black African group reporting higher levels of stigma compared with those with moderate ID. Younger age was the only factor that was associated with stigma but there was a trend towards ethnicity, additional disability and socio-economic status being related to stigma. Conclusion Interventions should target the Black African community in South Africa and should include the reduction of both public stigma and self-reported stigma. Keywords discrimination, ethnicity, intellectual disability, negative attitudes, stigma
Introduction Conceptualising stigma People with intellectual disability (ID) are one of the most stigmatised and excluded groups in society (European Union Monitoring and Advocacy Program 2005) and are more likely to be subjected to physical violence compared the general population (Hughes et al. 2012). The conceptualisation of
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
Journal of Intellectual Disability Research
volume 59 part 6 june 2015
531 A. Ali et al. • Ethnicity and stigma in intellectual disability
public or social stigma has evolved over the last five decades. Goffman’s (1963) characterisation of stigma as an attribute that is deeply discredited by a person’s society, resulting in his/her rejection and devaluation, has been argued by many researchers to have limited utility in understanding healthrelated stigma because of its outdated use of language, generalised application to a wide range of situations and focus on social interactions rather than political or structural forces (Weiss et al. 2006). Stigma is now considered to be influenced by structural and cultural forces. Link & Phelan (2001) went on to describe stigma as a process defined by labelling, stereotyping, separating, status loss and discrimination, occurring in the presence of an imbalance of political and economic power. The social cognitive model of stigma, formulated by Corrigan and colleagues, suggests that stigma comprises of cognitive, emotional and behavioural aspects, manifesting as stereotypes, prejudice and discrimination (Corrigan 2000; Corrigan & Watson 2002). At the individual level, stigma may become internalised (also known as self or felt stigma), which occurs when the individual endorses cultural stereotypes about their condition, applies it to him/ herself and believes that they will be devalued by others (Corrigan et al. 2006).
analysis of results from a large population-based study of 1273 participants with ID (aged over 16) residing in at home. Questions were asked about whether they had experienced bullying at school and whether people had been rude or nasty to them in the last 12 months. Half the sample reported that they had been bullied at school and a third reported that they had been the recipient of bullying in the last 12 months. Participants with ID have also reported discrimination in finding employment and from colleagues at work because of misconceptions about abilities (Li 2004). Jahoda & Markova (2004) found that participants who had recently moved from institutions into community settings, deliberately concealed their past in order to be accepted. Living independently was perceived as a positive step in overcoming stigma. The experience of stigma, particularly if it has been internalised, has been linked to poor psychological well-being in people with ID. Higher levels of reported stigma is associated with lower selfesteem (Szivos-Bach 1993; Abraham et al. 2002; Paterson et al. 2012), comparing oneself less favourable to others in the community (Dagnan & Waring 2004; Cooney et al. 2006; Paterson et al. 2012) and psychological distress (Emerson 2010; Paterson et al. 2012).
The experience of stigma by people with ID
Public’s knowledge and attitudes towards ID
Studies show that people with ID are aware of stigma and that they are viewed negatively by society. Cooney et al. (2006) examined whether there were differences in the levels of stigma experienced by students attending segregated schools and mainstream schools. The students attending mainstream schools were more likely to report stigmatising treatment compared with those attending segregated schools. However, both groups reported similar experiences outside the school environment. Chen & Shu (2012) found that students with ID attending a mainstream school in Taiwan reported feelings of shame and embarrassment as a result of ‘odd stares’ and ‘strange looks’ looks from their peers. The high value placed on academic ability in Taiwan and similar cultures may result in people with ID being regarded more negatively compared with other cultures that place less emphasis on education. Emerson (2010) conducted a secondary
The experience of stigma and discrimination by marginalised groups is closely related to public stigma, for example higher levels of stigma are reported by people with mental illness in countries with higher levels of public stigma towards mental illness (Vogel et al. 2013). It is therefore helpful to examine the literature in relation to public stigma towards people with ID. Scior (2011) conducted a systematic review, which found that negative attitudes towards people with ID were apparent across different cultures. Several studies consistently demonstrate that members of the lay public have a limited understanding of the concept of ID (Gordon et al. 2004), and that awareness and knowledge varies between respondents from different cultural and ethnic backgrounds. Knowledge about the prevalence of ID is often inaccurate (Alem et al. 1999; Tachibana & Watanabe 2003; Tachibana 2006). One study found that only one in
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 6 june 2015
Journal of Intellectual Disability Research 532 A. Ali et al. • Ethnicity and stigma in intellectual disability
four people were able to recognise a person with ID from a vignette. Recognition was highest among White British participants compared with those from Asian and African backgrounds (Scior & Furnham 2011). There are misconceptions about the aetiology of ID. In India, common explanations are that it was ‘God’s will’ or the fault of the parents (Madhavan et al. 1990) and in Tanzania, witchcraft was commonly cited as a cause by tribal leaders (Kisanji 1995). Studies examining social distance suggest that the general public are unlikely to consider someone with ID as a friend (Gordon et al. 2004). Attitudes towards people with more severe ID tend to be more negative compared with those with mild ID (Antonak et al. 1995) and are linked to perceptions about the ability and competence of people with ID. Attitudes in developed countries generally promote the inclusion of people with ID in society (Bryant et al. 2006; Henry et al. 1996). Negative attitudes are more likely to be associated with supernatural forces (Mulatu 1999) and views about the condition being self-inflicted (Panek & Jungers 2008). A few studies of cross-cultural comparisons suggest that some cultures may hold particularly negative attitudes. For example, Asian American students held more negative attitudes than Latin American or African students (Saetermore et al. 2001) and White British respondents were more positive towards people with ID compared with respondents from Hong Kong (Scior et al. 2010).The most consistent factor associated with positive attitudes is having prior contact with someone with ID. In particular, positive attitudes are associated with having positive contact (Morin et al. 2013).
Variables associated stigma and double stigma Studies of people with ID have not found a relationship between stigma and age (Szivos-Bach 1993; Cooney et al. 2006; Paterson et al. 2012), stigma and gender (Szivos-Bach 1993; Petrovski & Gleeson 1997; Dagnan & Waring 2004; Cooney et al. 2006; Paterson et al. 2012) or between stigma and IQ (Szivos-Bach 1993; Cooney et al. 2006; Paterson et al. 2012) possibly because the studies were not sufficiently powered to detect differences (sample size of the studies were 50 participants or less).
There is little research on the relationship between ethnicity, culture and the experience of stigma experienced by people with ID. In other areas of research, cultural factors have been found to influence the experience of stigma. For example one study of older adults found that Asian Americans and Latinos expressed greater shame and embarrassment about having mental illness compared with non-Latino Whites (Jimenez et al. 2013). Depressed women of immigrant African and Caribbean backgrounds, and USA-born Black or Latino women, were less likely to seek treatment compared with USA-born White women because of the fear of stigma (Nadeem et al. 2007). People with ID may have other potentially stigmatising features such as belonging to a non-White ethnic minority group, the presence of mental illness or behavioural problems, mobility and speech problems. The presence and combined effects of more than one stigmatising condition is known as ‘double stigma’, and may lead to individuals reporting higher levels of stigma compared with those who have just one stigmatising condition. There is relatively little research on the impact of multiple stigmatising characteristics such as the impact of ethnicity or racism and ID on the experience of stigma. However, one qualitative study of South Asians with ID living in the UK, suggests that the combined effects of racism and the presence of ID contributed to higher perceptions of discrimination and harassment (Azmi et al. 1997). In other areas of research, a higher perception of racial discrimination among Black women in Canada, has been correlated with higher HIV stigma (Logie et al. 2013).
Rationale for study Given that there is evidence for cultural differences in attitudes towards people with ID, one might expect there to be cultural differences in the experience of stigma by people with ID. However, there is little research in this area and no cross comparisons have been made examining the experience of stigma in people with ID from different communities. As there is relatively little known about ethnic or cultural differences in the experience of stigma, our study was aimed at examining differences in the reported experiences of stigma in people with ID from three ethnic groups in Cape
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 6 june 2015
Journal of Intellectual Disability Research 533 A. Ali et al. • Ethnicity and stigma in intellectual disability
Town, South Africa. Given the historical context of Apartheid South Africa, this has led to ethnic groups developing distinct cultural identities, which makes this area a unique location to study differences in stigmatisation.
Objectives 1 To identify whether there are differences in levels of stigma reported by three main ethnic groups in Cape Town: Xhosa-speaking Black Africans; those of mixed ancestry; and Afrikaans or Englishspeaking Caucasians of White European Descent; 2 To identify whether there are differences in selfreported stigma between those with mild or moderate ID, and between participants from other potentially stigmatising groups (e.g. those with physical disability, an unusual appearance, mental illness or behavioural problems); and 3 The assess for interaction effects between ethnicity and level of ID on self-reported stigma.
Hypotheses 1 Levels of self-reported stigma will differ between the three ethnic groups. The Xhosa-speaking Black African group will report higher levels of stigma because of the combined effects of potential racial discrimination and ID. 2 Levels of reported stigma will be higher in those with moderate ID compared with mild ID because the moderate group may have more ‘noticeable disability’. 3 Levels of self-reported stigma will also be higher in other groups with potentially stigmatising attributes. 4 The effect of ethnicity on self-reported stigma will be modified by level of ID.
Method Ethical approval Ethical approval was obtained from the Research Ethics Committee, Health Science Faculty, University of Cape Town. The study was conducted in the Western Cape Metropolitan Area, South Africa between August 2007 and May 2008.
Study area The study area was the metropolitan area of Cape Town in the Western Cape Province, South Africa. Cape Town has 3.4 million inhabitants, and includes the city of Cape Town, Mitchells Plain and Khayelitsha. During the 2007 community survey, 44% of the population indicated their population group as mixed ancestry, 35% was Black African, 19% was Caucasian and 2% were Asian. Afrikaans was the most common home language (41% of households) followed by Xhosa (29%) and English (28%) (Statistics South Africa 2009). The age distribution reflected a young population with 27% of people aged 14 years or younger. Sixty-five per cent of the population was younger than 35 years of age. In 2007 3.5% of the total population of Cape Town had a physical and mental disability.
Participants Potential participants with mild or moderate ID were identified by occupational therapists, supervisors, nurses or social workers attached to various sheltered employment workshops, outpatient clinics, the provincial government or local voluntary organisations. Participants were recruited based on language as well as ethnic group to ensure that we had approximately equal numbers of participants per language and ethnic group. Background information was gathered from the participant, as well as their caregiver or other informant (e.g. the occupational therapist at the sheltered employment, or the psychiatric nurse at the outpatient clinic). Participants were interviewed at their workshops, homes or clinics. Adults with mild to moderate ID were included. Participants under the age of 18 were excluded from this study as the stigma measure (see below) had not been validated in this group, and those who were unable to provide informed consent were also excluded. One hundred and ninety-one participants were included in the study. Fifty-three (27.7%) were Black Africans; 68 (35.6%) were Caucasians; and 70 (36.6%) were of mixed ancestry. Women made up 55.5% of the sample. One hundred and six (55.5%) had a mild ID and the remainder had a moderate ID. Seventy-one (37%) spoke Afrikaans, 67 (34.9%) English and 51 (27.6%) Xhosa.
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 6 june 2015
Journal of Intellectual Disability Research 534 A. Ali et al. • Ethnicity and stigma in intellectual disability
Measures Self-reported experiences of stigma Self-reported experiences of stigma were measured with the South African version of the Perceived Stigma of Intellectual Disability tool (Kock et al. 2012) that was originally developed in the UK (Ali et al. 2008). The UK version was translated into English, Afrikaans and Xhosa by a team of experts who worked with adults with ID. Independent translators back translated the measure into English to ensure that the items retained their original meanings. The face validity of the stigma measure was assessed by participants with ID in three focus groups representing the three languages. The measure was then piloted by two participants from each language (see Kock et al. 2012 for further information about the development of the measure). The South African stigma measure is a self-report instrument with two sub-scales: ‘felt stigma’ and ‘reaction to felt stigma’ and each item has an accompanying picture, with a yes/no response format. Items from the ‘felt stigma sub-scale include: ‘I think I am the same as other people’, ‘people on the street make fun of me’, ‘people make fun of me for going to the day centre’. Items from the ‘reaction to felt stigma sub-scale’ include: ‘the way people talk to me makes me angry’ and ‘people make me feel embarrassed’. In order to ensure that the responses given were valid, the interviewer asked the participant to give examples of situations when the item under question occurred. The items are scored 0–10. The tool has good psychometric properties (per item test re-test reliability kappa 0.41 to 0.59; factor loadings 0.51–0.83; Cronbach’s alpha 0.73).
behaviour drew attention from their community. In addition, socio-economic status was recorded, which was defined by area of residence based on data obtained from the 2001 census (Statistics South Africa 2003; Gie & Ramonovsky 2006). We rated socio-economic status on a five-point scale (1–5) with a score of one indicating higher socioeconomic status.
Analysis Analysis of data was performed using SPSS, version 17. In order to simplify the analysis, the variables relating to stigmatising features (apart from level of ID) were combined into two composite variables: ‘obvious additional disability’, and ‘behaviour or mental disorder’ and the responses were dichotomised to ‘yes’ and ‘no’. The demographic characteristics of the whole sample and the three ethnic groups were analysed using descriptive statistics. Individual stigma items were analysed to assess whether there were differences in the responses given by the different ethnic groups. As the total stigma score was normally distributed, one-way analysis of variance (anova) was used to test whether there were statistically significant differences in the total stigma score between the three ethnic groups. Interaction effects between ethnic group and the demographic and clinical variables were assessed using two-way between-group analysis of variance. A multiple linear regression analysis was performed to identify factors associated with stigma and to adjust for variables that were found to confound the relationship between ethnicity and the total stigma score.
Results Socio-demographic and clinical variables
Sample characteristics
A structured data collection form was used to record information on gender, age, ethnicity, language (Afrikaans, English and Xhosa), level of ID (which was based on a clinical assessment using International Classification of Disease (ICD)-10 criteria (World Health Organisation 1993), sensory problems, mobility problems, involuntary movements, physical illness, mental illness (e.g. epilepsy), medication, whether the individual looked noticeably different from other people and whether their
Demographic factors, according to ethnic group are shown in Table 1. The Black African group had a significantly higher proportion of people with mild ID (Chi 22.9, P < 0.001) and almost all the participants were living in areas with the lowest socio-economic ratings, compared with the Caucasian group, where the majority of the participants were living in areas with the highest socio-economic ratings (Chi 296.1, P < 0.001). In addition the majority of the Black African group lived within the
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 6 june 2015
Journal of Intellectual Disability Research 535 A. Ali et al. • Ethnicity and stigma in intellectual disability
Total number Gender: males Age (mean) Severity of ID Mild ID Moderate ID Language Xhosa Afrikaans English Socio-economic group 1 2 3 4 5 Accommodation Own home Family home Group home Foster care Mobility problems Unusual appearance Speech problems Behaviour or mental disorder
Black African Number (%) 53
Mixed race Number (%) 70
Caucasian Number (%) 68
24 (45.3) 25.7
30 (42.9) 32.7
31 (45.6) 34.7
40 (75.5) 13 (24.5)
39 (55.7) 29 (41.4)
27 (39.7) 31 (45.6)
53 (100) 0 (0) 0
0 (0) 52 (74.3) 18 (25.7)
0 (0) 19 (27.9) 49 (72.1)
0 0 0 1 (1.9) 52 (98.1)
3 (4.3) 6 (8.6) 24 (34.3) 37 (52.9) 0
38 (56.7) 22 (32.8) 2 (3.0) 4 (6.0) 1 (1.5)
0 52 (98.1) 1 (1.9) 0 2 (3.8) 8 (15.1) 2 (3.8) 12 (22.6)
44 (62.9) 25 (35.7) 1 (1.4) 14 (20.0) 22 (31.4) 19 (27.1) 33 (47.1)
1 (1.5) 34 (50.0) 33 (48.5) 0 7 (10.3) 27 (39.7) 14 (20.6) 41 (60.3)
family home, compared with other groups (Chi 35.92, P < 0.001). In general, the Black African group had fewer additional stigmatising features (e.g. mobility problems, speech and behavioural problems, unusual appearance, mental illness and epilepsy).
Distribution of individual stigma items and total scores across the three groups Only three items were scored significantly differently between the three ethnic groups (see Table 2). The black African group were more likely than the other two ethnic groups to report that they had been attacked by people on the street and that they had been stared at. However, they were also more likely to report that they thought they were the same as other people. The total stigma scores did not differ between the three groups (F = 0.62, P = 0.54). Analysis of interaction effects on level of stigma revealed that there was a significant interaction
Table 1 Socio-demographic characteristics of the three ethnic groups
between ethnic group and level of ID (F = 3.74, P = 0.01). A graphical plot (Fig. 1) revealed that participants with mild ID from the Black African group reported higher levels of stigma than those with moderate ID and this effect was less marked for the Caucasian group. Those with moderate ID from the mixed ethnic group were more likely to report higher levels of stigma than those with mild ID, although the difference in the scores is not as striking as the Black ethnic group.
Factors associated with stigma Table 3 shows the results of the regression analysis, with total stigma score as the dependent variable. The independent variables included were age, gender, ethnicity, level of ID, socio-economic status, the presence of a behaviour or mental disorder, and obvious additional disability. Lower age was significantly associated with stigma (P = 0.02); however, there was a strong trend towards an
© 2014 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and John Wiley & Sons Ltd
volume 59 part 6 june 2015
Journal of Intellectual Disability Research 536 A. Ali et al. • Ethnicity and stigma in intellectual disability
Table 2 Responses to individual items on the stigma instrument given by the three ethnic groups
Item
Black Number (%)
Mixed Number (%)
White Number (%)
Chi square (P value)
I think I am the same as other people The way people talk to me makes me angry People make me feel embarrassed People on the street make fun of me People on the street have hurt me People on the street look at me in a funny way People treat me like a child People laugh at me because of the way I talk I worry about the way people act towards me People make fun of me about going to the day centre
47 (88.7) 34 (64.2) 31 (58.5) 25 (47.2) 24 (45.3) 33 (62.3) 24 (46.2) 24 (46.2) 25 (48.1) 18 (34.0)
35 (50.0) 48 (68.6) 38 (54.3) 34 (54.3) 8 (11.4) 32 (45.7) 26 (37.1) 26 (37.1) 28 (40.0) 22 (31.4)
34 (50.0) 49 (72.1) 57.4 (39) 24 (35.3) 8 (11.8) 27 (39.7) 25 (36.8) 25 (36.8) 25 (36.8) 17 (25.0)
24.02 (
