Story · Violence
Untouchability in the data
How often do landlords and their agents refuse Dalit home-seekers, and how many households say untouchability is practised in their homes?
Key takeaways
- In person, 32 of 66 Dalit home-seekers (48.48%) got a positive answer from landlords or their agents in five cities around Delhi in 2012, against 64 of 66 dominant-caste Hindu home-seekers (96.97%). Exhibit 1
- By telephone, 58.62% of 493 Dalit callers got a positive answer, against 99.80% of 493 dominant-caste Hindu callers. Exhibit 2
- In the 2016 SARI phone survey, 39% of non-Dalit Hindu women and 27% of men in Delhi said someone in their household practises untouchability. Exhibit 3
- Of the 42,152 households in the IHDS-II sample across India, 2011-12, 27% said they practise untouchability in some form, 30% in rural areas and 20% in urban areas, a peer-reviewed study of IHDS-II data found. Exhibit 4
What an audit by researchers and two peer-reviewed studies found. A 2012 audit of the rental housing market in five cities of the National Capital Region (NCR) of Delhi (Delhi, Faridabad, Ghaziabad, Gurgaon and Noida), in which matched home-seekers whose names signalled their caste or religion asked landlords and agents about flats to rent, by telephone and in person. And the Social Attitudes Research, India (SARI) phone survey of 2016, which asked non-Dalit Hindus in Delhi, Mumbai, Rajasthan and Uttar Pradesh whether anyone in their household practises untouchability, beside the India Human Development Survey (IHDS) of 2011-12 for the same places. And a peer-reviewed study of the 42,152 households in the second round of the IHDS, 2011-12, across India, by rural and urban areas and by social group. Untouchability is the exclusion of Dalits as ritually polluting, abolished by Article 17 of the Constitution. Each figure covers its own sample, place and year; none is a national share.
- of 66 Dalit home-seekers got a positive answer from landlords or their agents in person, against 96.97% of 66 dominant-caste Hindu home-seekers (survey: an audit by researchers, five NCR cities, 2012)
- 48.48%
- of non-Dalit Hindu women in Delhi said someone in their household practises untouchability (peer-reviewed study, SARI phone survey, 2016)
- 39%
- of the 42,152 households in the IHDS-II sample across India, 2011-12, said they practise untouchability in some form (peer-reviewed study of IHDS-II data)
- 27%
Exhibit 1
In the 2012 in-person audit in five cities around Delhi, 32 of 66 Dalit home-seekers (48.48%) got a positive answer from landlords or their agents, against 64 of 66 dominant-caste Hindu home-seekers (96.97%)
Answers to matched home-seekers who met landlords, real estate agents and brokers in person, by the home-seeker’s caste or religion, Delhi, Faridabad, Ghaziabad, Gurgaon and Noida, January–March 2012. Cases, and per cent of each group, as Table 2 prints them.
Table: Dominant-caste Hindus, of 66: positive 64 (96.97%), positive on different terms 2 (3.03%), refused printed as a dash; Dalits, of 66: positive 32 (48.48%), positive on different terms 5 (7.58%), refused 29 (43.94%); Muslims, of 66: positive 19 (28.79%), positive on different terms 7 (10.60%), refused 40 (60.61%).
Home-seekers Positive answer Positive Positive, on different terms On different terms Refused Refused All Dominant-caste Hindus 64 96.97% 2 3.03% – – 66 Dalits 32 48.48% 5 7.58% 29 43.94% 66 Muslims 19 28.79% 7 10.60% 40 60.61% 66 Note: A small sample: 66 home-seekers in each group, 198 in all. The authors say the higher share of refusals in person may also be due to the small sample. A dash for Dominant-caste Hindus’ refusals is as the table prints it; that group’s other two rows add up to all 66. “Dominant-caste Hindus” are the study’s “upper-caste Hindus”.
Source (survey): Thorat S, Banerjee A, Mishra VK and Rizvi F, “Urban Rental Housing Market: Caste and Religion Matters in Access”, Economic & Political Weekly, Vol 50, Nos 26 and 27, pp 47-53, dated 27 June 2015, Table 2. A sample, not a national count. Analysis by The Dalit Project, as on 28 Oct 2026. Exhibit 2
By telephone, 289 of 493 Dalit callers (58.62%) got a positive answer, against 492 of 493 dominant-caste Hindu callers (99.80%)
Answers to matched home-seekers who telephoned the landlords, agents and brokers behind advertised flats, by the caller’s caste or religion, Delhi, Faridabad, Ghaziabad, Gurgaon and Noida, January–March 2012. Cases, and per cent of each group, as Table 1 prints them.
Table: Dominant-caste Hindus, of 493: positive 492 (99.80%), positive on different terms 1 (0.20%), refused 0 (0.00%); Dalits, of 493: positive 289 (58.62%), positive on different terms 114 (23.12%), refused 90 (18.26%); Muslims, of 493: positive 165 (33.47%), positive on different terms 176 (35.70%), refused 152 (30.83%).
Home-seekers Positive answer Positive Positive, on different terms On different terms Refused Refused All Dominant-caste Hindus 492 99.80% 1 0.20% 0 0.00% 493 Dalits 289 58.62% 114 23.12% 90 18.26% 493 Muslims 165 33.47% 176 35.70% 152 30.83% 493 Note: The callers differed only in a name chosen to show their caste or religion. The paper’s text gives other figures from its table: “While all the 493 high-caste home-seekers received a positive response, the corresponding proportion fell to 59% for the Dalits and to merely 29% for the Muslims”. The table is used here. “Dominant-caste Hindus” are the study’s “upper-caste Hindus”.
Source (survey): Thorat S, Banerjee A, Mishra VK and Rizvi F, “Urban Rental Housing Market: Caste and Religion Matters in Access”, Economic & Political Weekly, Vol 50, Nos 26 and 27, pp 47-53, dated 27 June 2015, Table 1. A sample, not a national count. Analysis by The Dalit Project, as on 28 Oct 2026. Exhibit 3
Asked in 2016 whether anyone in their household practises untouchability, 21% of non-Dalit Hindu men in Mumbai said yes, and 66% of women in rural Rajasthan
Non-Dalit Hindus who said they or someone in their household practise untouchability, %, by place: the SARI phone survey (2016), and the IHDS (2011-12) for the same places, as Table 3 of the study prints them.
Table: Rural Rajasthan: SARI 2016, women: someone in the household 66%, SARI 2016, men: someone in the household 50%, IHDS 2011-12: someone in the household 50%, IHDS 2011-12: household, or the kitchen question 68%, SARI 2016, women: themselves (asked after a yes) 54%, SARI 2016, men: themselves (asked after a yes) 34%; Rural Uttar Pradesh: SARI 2016, women: someone in the household 64%, SARI 2016, men: someone in the household 43%, IHDS 2011-12: someone in the household 46%, IHDS 2011-12: household, or the kitchen question 56%, SARI 2016, women: themselves (asked after a yes) 48%, SARI 2016, men: themselves (asked after a yes) 28%; Urban Rajasthan: SARI 2016, women: someone in the household 50%, SARI 2016, men: someone in the household 33%, IHDS 2011-12: someone in the household 44%, IHDS 2011-12: household, or the kitchen question 47%, SARI 2016, women: themselves (asked after a yes) 31%, SARI 2016, men: themselves (asked after a yes) 15%; Urban Uttar Pradesh: SARI 2016, women: someone in the household 48%, SARI 2016, men: someone in the household 42%, IHDS 2011-12: someone in the household 40%, IHDS 2011-12: household, or the kitchen question 51%, SARI 2016, women: themselves (asked after a yes) 35%, SARI 2016, men: themselves (asked after a yes) 18%; Delhi: SARI 2016, women: someone in the household 39%, SARI 2016, men: someone in the household 27%, IHDS 2011-12: someone in the household 12%, IHDS 2011-12: household, or the kitchen question 20%, SARI 2016, women: themselves (asked after a yes) 21%, SARI 2016, men: themselves (asked after a yes) 12%; Mumbai: SARI 2016, women: someone in the household not asked, SARI 2016, men: someone in the household 21%, IHDS 2011-12: someone in the household 5%, IHDS 2011-12: household, or the kitchen question 5%, SARI 2016, women: themselves (asked after a yes) not asked, SARI 2016, men: themselves (asked after a yes) 10%.
Place SARI 2016, women: someone in the household SARI 2016, men: someone in the household IHDS 2011-12: someone in the household IHDS 2011-12: household, or the kitchen question SARI 2016, women: themselves (asked after a yes) SARI 2016, men: themselves (asked after a yes) Rural Rajasthan 66% 50% 50% 68% 54% 34% Rural Uttar Pradesh 64% 43% 46% 56% 48% 28% Urban Rajasthan 50% 33% 44% 47% 31% 15% Urban Uttar Pradesh 48% 42% 40% 51% 35% 18% Delhi 39% 27% 12% 20% 21% 12% Mumbai – 21% 5% 5% – 10% Note: Self-reported: these are what people told surveyors, so anyone who practises untouchability and does not say so is not counted. SARI asked women and men separately, and in Mumbai men only; in the IHDS, “Almost 90% of the primary respondents to the IHDS household questionnaire, where answers to untouchability questions were recorded, were adult women”. SARI asked by phone and the IHDS face to face; the study calls even its closest pair, SARI women and the IHDS with the kitchen question, “not directly comparable”. The “themselves” columns answer a question SARI asked only those who said yes about their household; the study does not say whether those per cents are of everyone it asked or only of those who said yes.
Source (peer-reviewed study): Coffey D, Hathi P, Khurana N and Thorat A, “Explicit Prejudice: Evidence from a New Survey”, Economic & Political Weekly, Vol 53, No 1, pp 46-54, dated 6 January 2018, Table 3. A sample, not a national count. The IHDS columns are the study authors’ figures from IHDS 2011-12 data. Analysis by The Dalit Project, as on 28 Oct 2026. Exhibit 4
Of the 42,152 households in the IHDS-II sample across India, 2011-12, 27% said they practise untouchability in some form: 30% in rural areas and 20% in urban areas
Households that said yes to either of the IHDS-II questions on untouchability, %, all India, 2011-12, as the study’s text states them.
Table: all households in the sample, 27%; rural areas, 30%; urban areas, 20%.
Households Said they practise untouchability All households in the sample 27% Rural areas 30% Urban areas 20% Note: Self-reported, by the household’s primary respondent: a household counts if the answer was yes to “Do some members practise untouchability in your household?” or, after a no, to “Would there be a problem if someone who is a Scheduled Caste were to enter your kitchen or share utensils?” Households that practise untouchability and did not say so are not counted. The paper’s Figure 1 shows the rural and urban shares; the copy read here, a ProQuest print of the text, does not reproduce it, and the figures are taken from the text.
Source (peer-reviewed study): Thorat A and Joshi O, “The Continuing Practice of Untouchability in India: Patterns and Mitigating Influences”, Economic & Political Weekly, Vol 55, No 2, dated 11 January 2020, from its text. A sample, not a national count. The figures are from the study authors’ reading of IHDS-II data. Analysis by The Dalit Project, as on 28 Oct 2026. Exhibit 5
In the same sample, 22% of Scheduled Caste households, 15% of Scheduled Tribe households, 24% of forward-caste households and 52% of Brahmin households said they practise untouchability in some form
Households that said yes to either of the IHDS-II questions on untouchability, %, by the household’s social group, all India, 2011-12, as the study’s text states them, in the order the study lists the groups.
Table: Scheduled Castes (SCs), 22%; Scheduled Tribes (STs), 15%; Forward castes, 24%; Brahmins, 52%.
Social group Said they practise untouchability Scheduled Castes (SCs) 22% Scheduled Tribes (STs) 15% Forward castes 24% Brahmins 52% Note: The share for Other Backward Classes (OBCs) is in the paper’s Figure 2, which the copy read here, a ProQuest print of the text, does not reproduce, so it is not shown. The study splits the general category into Brahmins and “forward castes”, which it describes as Kshatriyas and Vaishyas; households that named no group are not shown either. Self-reported, by the household’s primary respondent. Groups are not ranked.
Source (peer-reviewed study): Thorat A and Joshi O, “The Continuing Practice of Untouchability in India: Patterns and Mitigating Influences”, Economic & Political Weekly, Vol 55, No 2, dated 11 January 2020, from its text. A sample, not a national count. The figures are from the study authors’ reading of IHDS-II data. Analysis by The Dalit Project, as on 28 Oct 2026.
What’s next
The next update
We will update these figures on release of a new study or survey that counts untouchability or audits rental housing. This is version 1.
What the data can’t tell us yet
A national share: each figure covers its study’s own sample, place and year, and the survey figures count only those who said so.
What you can do
- Cite this piece.
- Found an error? Tell us how to correct it.
About this data
- What this shows
- How landlords and agents in five cities around Delhi answered matched Dalit, Muslim and dominant-caste Hindu home-seekers in 2012, how many non-Dalit Hindus in four places said in 2016 and 2011-12 that untouchability is practised in their households, and how many households in the IHDS-II sample across India said so in 2011-12, by rural and urban areas and by social group.
- Source
- Economic & Political Weekly, Vol 50, Nos 26 and 27, pp 47-53; Economic & Political Weekly, Vol 53, No 1, pp 46-54; India Human Development Survey (IHDS); Economic & Political Weekly, Vol 55, No 2, 4 documents, Table 1 and Table 2 and Table 3. Analysis by The Dalit Project, as on 28 Oct 2026. Every source
- What it does not show
- A national share: each figure covers its study’s own sample, place and year, and the survey figures count only those who said so.
- Last updated
- 28 Oct 2026 · Next expected update: on release of a new study or survey that counts untouchability or audits rental housing
- Corrections
- Changelog (version 1) · Corrections policy