4. Ethnic Data Collection: Benefits, Risks, Data Sources and Methods

Chapter 3 entitled ‘Ethnic data collection: key elements, rules and principles’.

This was stressed in the introduction to the operational and organisational principles for sensitive data processing in Chap. 3 (Sect. 3.​7). Makkonen (2010), p. 227.

See Chap. 3 (Sect. 3.​7.​5) on active and meaningful involvement and awareness-raising as an important organisational principle for ethnic data collection and processing.

Chapter 5 entitled ‘Challenges to collecting ethnic data on the Roma minority in Europe’.

Report of the Independent Expert on Minority Issues on her Mission to Hungary (4 January 2007), para. 100. Lamberts et al. (2014), p. 32. Office of the High Commissioner for Human Rights (OHCHR) (2012), p. 68. Waldron (2011), p. 49. Covrig (2004), pp. 94, 99 and 100.

See, for example: European Commission against Racism and Intolerance (ECRI), Fifth Report on Belgium (4 December 2013), paras. 121 and 122. ECRI, Fourth Report on France (29 April 2010), para. 153. Wrench (2013), p. 79. Waldron (2011), p. 48. Ringelheim (2011), p. 1693. Wrench (2011), p. 1715. Makkonen (2006), p. 43. Ramsay (2006), pp. 1, 2 and 7. Negrin (2003). Haug (2001), p. 303. Official statistics will be considered as the first ethnic data source in Sect. 4.3.1.

European Network Against Racism (ENAR) (2012), p. 9. Makkonen (2010), p. 218. World Health Organization (WHO Europe) (2010), p. 8. Makkonen (2006), p. 5. See Sect. 4.1.4 on policy implementation and evaluation as the fourth benefit of ethnic data collection. See also Chap. 6 (Sect. 6.​3.​4), where the availability of ethnic data is cited as a prerequisite of positive action.

United Nations Statistics Division (UNSD) (2008b), para. 2.160.

Id. at para. 3.111.

Ethnic data can provide information on the situation of minority groups within one country as well as across national borders. United Nations Development Programme (UNDP) (2010), p. 95. UNSD (2008b), para. 2.160.

Oppenheimer (2008), pp. 738 and 739.

Recognition can be directly (Constitution, public policy or legal registration) or indirectly (census or survey). Committee on the Elimination of Racial Discrimination (CERD Committee), General Recommendation No. 24: Article 12 of the Convention Women and Health (27 August 1999), para. 3. Möschel (2009), p. 204.

See Chap. 2 on special measures to achieve full and effective equality for minorities (Sect. 2.​4.​2) and on the right to identity and the prohibition of assimilation as interrelated building blocks of minority rights protection (Sect. 2.​4.​3). UNSD (2003), p. 2.

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 31. Report of the Special Rapporteur on racism, racial discrimination, xenophobia and related forms of intolerance: Follow-up to and implementation of the Durban Declaration and Programme of Action (19 May 2009), para. 20. See Chap. 3 (Sect. 3.​2.​2) on how to determine the existence of ethnic minorities. Specifically in relation to Roma, the absence of uniform status of this minority throughout Europe was discussed in Chap. 1 (Sect. 1.​1.​3). As will be explained in Chap. 5 (Sect. 5.​3.​2), official recognition as a minority does not guarantee that Roma are included in official data collection practices. Furthermore, States sometimes collect data on this population group even when they are not an officially recognised minority.

UNSD (2008b), para. 2.160. See Chap. 1 (Sect. 1.​2.​3) for an introduction to intersectional discrimination and its relevance in relation to Roma in Europe. The need for an intersectional approach to equality data collection was also stressed in Chap. 3 (Sect. 3.​1.​2) on the call for ethnic data collection for equality and anti-discrimination purposes by various actors at the international and European level.

See, for example: Report of the Independent Expert on Minority Issues (6 January 2014), paras. 53 and 54. Report of the Independent Expert on Minority Issues on the Implementation of General Assembly Resolution 60/251 of 15 March 2006 entitled “Human Rights Council” (2 February 2007), paras. 23 to 46. Report of the Special Rapporteur on racism, racial discrimination, xenophobia and related forms of intolerance: Follow-up to and implementation of the Durban Declaration and Programme of Action (19 May 2009), para. 17. Report of the Secretary-General on the Achievement of the Goal and Objectives of the Second International Decade of the World’s Indigenous People (6 August 2014), para. 48. The particular vulnerability of Roma as a result of the intersection of discrimination, poverty and gender was considered in Chap. 1 (Sect. 1.​2.​3). Poverty will be discussed further on in this chapter when considering the usefulness of proxies for ethnicity in Sect. 4.5.4.

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 2. ECRI, General Policy Recommendation No. 13: Combating Anti-Gypsyism and Discrimination against Roma (24 June 2011), para. 14. ECRI, General Policy Recommendation No. 10: Combating racism and racial discrimination in and through school education (15 December 2006), para. 1(b). ECRI, General Policy Recommendation No. 4: National surveys on the experience and perception of discrimination and racism from the point of view of potential victims (6 March 1998), paras. 1, 6 and 9. Commission Communication, Non-discrimination and equal opportunities: A renewed commitment (2 July 2008), p. 7. European Union Agency for Fundamental Rights (FRA) (2013), p. 179. Hermanin et al. (2013b), p. 4. OHCHR (2012), p. 68. Wrench (2011), p. 1715. Makkonen (2010) p. 212. Bond et al. (2010), p. 1. Landman and Carvalho (2010), pp. 49 and 50. FRA (2007), pp. 49 and 50. The lack of an explicit legal obligation to collect ethnic data at the level of the United Nations (UN), the Council of Europe and the European Union (EU) was mentioned in Chap. 3 (Sect. 3.​1.​1).

Commission Green Paper, Equality and non-discrimination in an enlarged European Union (28 May 2004), p. 17. Ringelheim and De Schutter (2010), Foreword. Ringelheim (2008/9), p. 50.

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 31. Commission Green Paper, Equality and non-discrimination in an enlarged European Union (28 May 2004), p. 17. ENAR (2012), p. 6. Makkonen (2010), p. 219. Ringelheim and De Schutter (2010), Foreword. Makkonen (2006), p. 42.

See Chap. 1 (Sect. 1.​3.​1) for a definition of the notion equality data. Commission Communication, Joint Report on the application of Council Directive 2000/43/EC of 29 June 2000 implementing the principle of equal treatment between persons irrespective of racial or ethnic origin (‘Racial Equality Directive’) and of Council Directive 2000/78/EC of 27 November 2000 establishing a general framework for equal treatment in employment and occupation (‘Employment Equality Directive’) (17 January 2014), p. 5.

Report of the Independent Expert on Minority Issues (6 January 2014), paras. 53, 54, 63 and 86. Report of the Special Rapporteur on contemporary forms of racism, racial discrimination, xenophobia and related intolerance on the implementation of Assembly resolution 67/154 (16 August 2013), para. 146. Guidance Note of the United Nations Secretary-General on Racial Discrimination and Protection of Minorities (March 2013), para. 23. Guidelines for the CERD-specific document to be submitted by States parties under article 9, paragraph 1, of the Convention (13 June 2008), paras. 3 and 12. Report of the Independent Expert on Minority Issues on the Implementation of General Assembly Resolution 60/251 of 15 March 2006 entitled “Human Rights Council” (2 February 2007), summary. Report of the Permanent Forum on Indigenous Issues on the fifth session (2006), paras. 6 and 36. Durban Declaration (8 September 2001), para. 92. Committee on the Rights of the Child (CRC Committee), General Comment No. 11: Indigenous children and their rights under the Convention (12 February 2009), paras. 24 and 26. CRC Committee, General Comment No. 9: The rights of children with disabilities (27 February 2007), para. 19. CRC Committee, General Comment No. 5: General measures of implementation of the Convention on the Rights of the Child (27 November 2003), paras. 9 and 12. Committee on Economic, Social and Cultural Rights (CESCR Committee), General Comment No. 13: The Right to Education (Art. 13) (8 December 1999), para. 37. CERD Committee, General Recommendation No. 31: The prevention of racial discrimination in the administration and functioning of the criminal justice system (20 August 2004), para. 1.2. UNDP (2010), pp. 3, 33 and 39. Lamberts et al. (2014), p. 10.

The MDGs were established in follow-up of the 2000 UN Millennium Summit and ended in 2015. Monitoring, evaluation and accountability were not mentioned in the MDGs. The MDGs, which were addressed to developing countries, were: (1) halving the number of undernourished people, (2) achieving universal primary education, (3) promoting gender equality and empowering women, (4) reducing child mortality, (5) improving maternal health, (6), combating HIV/aids, malaria and other diseases, (7) ensuring environmental sustainability, and (8) developing a global partnership for development. Only a few States presented ethnically disaggregated data in the MDG reports, though not systematically. United Nations Millennium Declaration (8 September 2000). Report of the Independent Expert on Minority Issues (6 January 2014), para. 36. Report of the Special Rapporteur on contemporary forms of racism, racial discrimination, xenophobia and related intolerance (19 August 2013), paras. 45 and 71. Report of the Permanent Forum on Indigenous Issues on the fifth session (2006), paras. 6, 36, 92 to 94 and 102. Report of the Independent Expert on Minority Issues on the Implementation of General Assembly Resolution 60/251 of 15 March 2006 entitled “Human Rights Council” (2 February 2007), paras. 68, 76 and 104(f). UN (2014), pp. 6 and 7. UNDP (2010), pp. 73, 75, 81 and 90. Ramsay (2006), pp. 5 and 6. Minority Rights Group International (2003), p. 1.

The 17 SDGs were adopted in September 2015 and entered into force on 1 January 2016. These global goals, which build on the MDGs and apply to all countries countries, are: (1) no poverty, (2) zero hunger, (3) good health and well-being, (4) quality education, (5) gender equality, (6) clean water and sanitation, (7) affordable and clean energy, (8) decent work and economic growth, (9) industry, innovation and infrastructure, (10) reduced inequalities, (11) sustainable cities and communities, (12) responsible consumption and reproduction, (13) climate action, (14) life below water, (15) life on land, (16) peace, justice and strong institutions, and (17) partnerships for the goals. Report of the Open Working Group of the General Assembly on Sustainable Development Goals (12 August 2014).

Negrin (2003).

European Roma Rights Centre (ERRC) (2013), p. 42.

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 2.

Report of the Special Rapporteur on contemporary forms of racism, racial discrimination, xenophobia and related intolerance (19 August 2013), para. 70. CERD Committee, General Recommendation No. 32: The meaning and scope of special measures in the International Convention on the Elimination of All Forms of Racial Discrimination (24 September 2009), para. 17. CERD Committee, General Recommendation No. 24: Article 12 of the Convention Women and Health (27 August 1999), paras. 3 and 4. Report of the Special Rapporteur on racism, racial discrimination, xenophobia and related forms of intolerance: Follow-up to and implementation of the Durban Declaration and Programme of Action (19 May 2009), paras. 21, 22 and 52. Durban Plan of Action (8 September 2001), para. 92.

Complaints data are, however, not the best indicator of levels of discrimination, as will be discussed in Sect. 4.3.4. Equality bodies and their role in ethnic data collection was briefly addressed in Chap. 3 (Sect. 3.​1.​1). Council Directive 2000/43/EC implementing the principle of equal treatment between persons irrespective of racial or ethnic origin (29 June 2000) (RED), art. 13. Muigai (2000), pp. 2 and 4. ENAR (2012), p. 8. FRA (2010c), p. 16.

For more on this, see Sect. 4.1.4 on policy implementation and evaluation as the fourth benefit of ethnic data collection. See also Chap. 6 for a discussion on the availability of ethnic data as a premise to render positive action successful (Sect. 6.​3.​4) and for an overview of the five main justifications and aims of positive action (Sect. 6.​4).

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 35.

Alidadi (2017), p. 17. Farkas (2017), p. 48. ENAR (2012), p. 8. Wrench (2011), p. 1716. UNDP (2010), pp. 33 and 103. Makkonen (2010), p. 219. Gray (2009), p. 55. Makkonen (2006), pp. 12 and 84. Hollo (2006), p. 24. Simon (2005), pp. 14 and 20. Cahn (2004), p. 36.

FRA (2007), p. 151. Hollo (2006), p. 24.

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 35.

Simon et al. (2015), p. 6.

Further on, political will is identified as the third key principle. See Chap. 5 (Sect. 5.​8) on awareness-raising, active participation, and genuine political will as three key principles of ethnic data collection on Roma in Europe, Chap. 6 on the need for active participation of all stakeholders in positive action (Sect. 6.​3.​2) and on the need for political will to implement effective positive action schemes (Sect. 6.​3.​3), and Chap. 9 on the lack of awareness about positive action and their relevance (Sect. 9.​2.​1), lack of political will to use positive action on Roma (Sect. 9.​2.​2) and the lack of consultation with and participation of local communities (Sect. 9.​2.​4) as challenges limiting the effectiveness of positive action for Roma, and Chap. 11 (Sect. 11.​5) on challenges to the effectiveness of inter-cultural mediation to enhance Roma inclusion. See also Chap. 12 (Sect. 12.​2) where awareness-raising, active participation and political will are cited as the three key elements of Roma equality and inclusion efforts in Europe.

ENAR (2012), pp. 1, 6, 8 and 10. Krizsán (2011), p. 1393. FRA (2009b), p. 19. Gray (2009), p. 62. Negrin (2003).

This requires regular data collection. Resolution 1740 of the Parliamentary Assembly on the situation of Roma in Europe and relevant activities of the Council of Europe (22 June 2010), para. 15.7. Resolution of the European Parliament on Non-Discrimination and Equal Opportunities for All—A Framework Strategy (14 June 2006), paras. 13 to 21 and recitals O to Q. Report of the Special Rapporteur on the right of everyone to the enjoyment of the highest attainable standard of health, Paul Hunt, on his Mission to Sweden (28 February 2007), para. 120. Commission Communication, Joint Report (17 January 2014), pp. 5 and 6. Commission Communication, Non-discrimination and equal opportunities: A renewed commitment (2 July 2008), p. 7. Commission Communication, Application of Directive 2000/43/EC of 29 June 2000 implementing the principle of equal treatment between persons irrespective of racial or ethnic origin (30 October 2006) p. 8. Alidadi (2017), p. 17. Wrench (2013), p. 79. ENAR (2012), p. 6. Waldron (2011), pp. 50 and 51. Makkonen (2010), p. 211. Ringelheim (2008/9), p. 41. Makkonen (2006), p. 84. Ramsay (2006), pp. 1 and 2. Simon (2005), p. 14.

This is particularly true for the Roma minority in Europe, as will be explained in Chap. 5 (Sect. 5.​3) on the heterogeneity among Roma and the challenges this poses to the construction of ethnic categories for data collection purposes. The diversity among Roma was previously also stressed in Chap. 1.

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 32. Report of the Special Rapporteur on racism, racial discrimination, xenophobia and related forms of intolerance: Follow-up to and implementation of the Durban Declaration and Programme of Action (19 May 2009), para. 22.

McDonald and Negrin (2010), p. 18. The importance of the availability of ethnically disaggregated data for the implementation and monitoring of special measures targeting vulnerable groups will also be underlined in Chap. 6 (Sect. 6.​3.​4) on positive action and the link with ethnic data.

Report of the Independent Expert on Minority Issues (6 January 2014), para. 86. UNSD (2003), p. 2.

See Part II of the book (Chaps. 6–11). The link between ethnic data and positive action will be explored in Chap. 6 (Sect. 6.​3.​4). Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 33. Guidance Note of the United Nations Secretary-General on Racial Discrimination and Protection of Minorities (March 2013), para. 23. CRC Committee, General Comment No. 11: Indigenous children and their rights under the Convention (12 February 2009), paras. 26 and 47. CERD Committee, General Recommendation No. 32: The meaning and scope of special measures in the International Convention on the Elimination of All Forms of Racial Discrimination (24 September 2009), paras. 17 and 37. Report of the Special Rapporteur on racism, racial discrimination, xenophobia and related forms of intolerance: Follow-up to and implementation of the Durban Declaration and Programme of Action (19 May 2009), para. 19. Report of the Independent Expert on Minority Issues on the Implementation of General Assembly Resolution 60/251 of 15 March 2006 entitled “Human Rights Council” (2 February 2007), para. 82 and 104(d)(e). Durban Plan of Action (8 September 2001), paras. 99 and 100. UNDP (2010), p. 37. Škobla et al. (2009), p. 27. Chopin et al. (2014), pp. 33 and 34. Hermanin (2013), pp. 104 and 105. Ringelheim (2011), p. 1683. Ringelheim and De Schutter (2010), p. 38. Cardinale (2007), p. 38. Makkonen (2006), pp. 12, 14 and 84. Rallu et al. (2006), p. 536. Ramsay (2006), p. 1.

States need ethnically disaggregated data to report to international and European monitoring bodies, as was explained in Chap. 3 (Sect. 3.​1.​2). The significance of international and European monitoring and litigation will be further explored in Chap. 5 (Sect. 5.​2.​4) on the multifarious data sources on Roma in Europe.

Hermanin and Atanasova (2013). Makkonen (2006), p. 84. Ramsay (2006), p. 2. Petrova (2004), p. 5.

Report of the Special Rapporteur on racism, racial discrimination, xenophobia and related forms of intolerance: Follow-up to and implementation of the Durban Declaration and Programme of Action (19 May 2009), para. 24 (a). Farkas (2011), p. 47. Makkonen (2010), p. 214. Muigai (2000), pp. 6 and 7. Ringelheim (2008/9), p. 56. Cardinale (2007), p. 38. Makkonen (2006), pp. 4, 12, 14 and 40. Simon (2007), p. 14. Ramsay (2006), p. 2. Simon (2005), p. 14. Petrova (2004), p. 5. Seltzer and Anderson (2001), pp. 484 and 506. This argument will be used in Sect. 4.2.1 to refute the fear that ethnic data may expose bad government policies.

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 31.

FRA (2010c), p. 25. Makkonen (2010), pp. 213 and 215. Ringelheim (2008/9), pp. 59 and 60. De Schutter and Ringelheim (2008), pp. 358, 376, 378, 382 and 383. Cardinale (2007), p. 38. Simon (2007), pp. 52 and 69. Ramsay (2006), p. 2. Deitch et al. (2003), pp. 1301, 1302, 1316 and 1317. Ethnic profiling will be discussed in Sect. 4.2.4 on the risks of ethnic data collection. See also Chap. 2 (Sect. 2.​1.​2) for a definition of direct and indirect discrimination and an exploration on the link with statistics and Chap. 3 (Sect. 3.​5.​3) on sensitive data processing for profiling purposes in the police sector.

Alidadi (2017), p. 17. See also Sect. 4.2.1 where it is argued that governments can rely on ethnic data to refute structural discrimination claims.

Resolution of the European Parliament on Non-Discrimination and Equal Opportunities for All—A Framework Strategy (14 June 2006), paras. 18 and 19 and recitals R and S. De Schutter and Ringelheim (2008), pp. 382 and 383. Makkonen (2006), pp. 32 and 33. Schiek (2007), pp. 397 and 398. Simon (2007), p. 69. Simon (2005), pp. 10, 13 and 14. Townshend-Smith (1995), p. 105. This was previously discussed in Chap. 2 (Sect. 2.​1.​2) on direct and indirect discrimination and the link with statistics.

RED, recital 21 and art. 8.1. Wrench (2011), p. 1717. De Schutter (2010), p. 26. Ringelheim (2006/7), p. 60.

Ethnic data can also establish a presumption of direct discrimination, though usually this type of discrimination does not require statistical reasoning, except when there is a pattern of direct discrimination against a specific group and there is only weak or no other evidence available. Member States such as Belgium, Czech Republic, France, Finland, Hungary, Italy, Sweden and the UK allow the use of statistics in legal proceedings. The use of ethnic statistics is however not as widespread as the use of gender and age in discrimination cases across the EU. RED, recital 15. Indirect discrimination and the link with data was first introduced in Chap. 2 (Sect. 2.​1.​2).

This case was mentioned and the importance of ethnic data was stressed in Chap. 2 (Sect. 2.​1.​2). See also Chap. 5 (Sect. 5.​2.​4) on the significance of international and European monitoring and litigation as a data source on Roma. European Court of Human Rights (EctHR), D H. and Others v. the Czech Republic, Judgment (13 November 2007), paras. 46, 137, 164, 180, 187 and 188. See also: EctHR, Oršuš and Others v. Croatia, Judgment (16 March 2010, GC), para. 152. Makkonen (2010), pp. 213 and 215. De Schutter and Ringelheim (2008), pp. 376, 383 and 384. Cardinale (2007), p. 38. Farkas (2007), p. 36. Ringelheim (2006/7), pp. 60 and 61. Hollo (2006), p. 48.

See, for instance: CJEU, Debra Allonby v. Accrington & Rossendale College, Education Lecturing Services, Judgment (13 January 2004), para. 81. CJEU, Hilde Schönheit v. Stadt Frankfurt am Main and Silvia Becker v. Land Hessen, Judgment (23 October 2003), paras. 67 to 69 and 71. CJEU, Seymour-Smith and Perez, paras. 49, 57, 59 to 65 and 77. Schiek (2007), p. 397.

Farkas (2011), p. 49. Makkonen (2010), p. 214. Hollo (2006), p. 48. As explained in Chap. 1 (Sect. 1.​3.​1) when introducing the main topics of this book and in Chap. 3 (Sect. 3.​1) on international and European views on ethnic data collection, such data are often unavailable. See also Chap. 5 (Sect. 5.​1) on large quantitative and qualitative data gaps on Roma communities in Europe.

Informal ethnic profiling concerns assumptions of law enforcement officers. For different definitions of ethnic or racial profiling, see, for example: ECRI, General Policy Recommendation No. 11: Combating racism and racial discrimination in policing (29 June 2007), para. 1. De Schutter and Ringelheim (2008), pp. 362 and 363. Discriminatory ethnic profiling was mentioned briefly in Chap. 3 (Sect. 3.​5.​3) on sensitive data processing for profiling purposes in the police sector.

Examples from ECRI country monitoring: ECRI, Fourth Report on Spain (7 December 2010), paras. 200 to 204. ECRI, Fourth Report on the United Kingdom (17 December 2009), para. 206. Recommendation CM/Rec(2010)13 of the Committee of Ministers to Member States on the protection of individuals with regard to automatic processing of personal data in the context of profiling (23 November 2010), para. 3.11. ECRI, General Policy Recommendation No. 11: Combating racism and racial discrimination in policing (29 June 2007), paras. 2, 36 and 41 to 43. For more on ethnic profiling as a possible risk of ethnic data collection, see Sect. 4.2.4.

Recommendation of the European Parliament to the Council of 24 April 2009 on the problem of profiling, notably on the basis of ethnicity, and race, in counter-terrorism, law enforcement, immigration, customs and border control (24 April 2009), para. W-AC. FRA (2010b), p. 1. FRA (2010c), pp. 5, 12, 13, 15, 18, 26, 37, 40, 41, 43 and 44.

FRA (2010c), p. 25. This will be addressed further in Sect. 4.2.4 on discriminatory ethnic profiling by public bodies as a risk of collecting ethnically disaggregated data. Suggested further reading on how to prevent unlawful profiling: FRA (2018).

This is due to a too restricted interpretation of EU data protection legislation. FRA (2010c), p. 25.

The five main benefits of ethnic data collection were presented in Sect. 4.1.

Report of the High Commissioner for Human Rights containing a draft basic document on the development of a racial equality index (31 January 2006), para. 71. Waldron (2011), p. 49. Jacobs et al. (2009), p. 68. Ringelheim (2006/7), pp. 53 and 54. Covrig (2004), pp. 93 and 94.

Eide (2013), pp. 49 and 50. Cahn (2004), p. 35.

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 40. Ramsay (2006), p. 2.

EctHR, D.H. and Others v. the Czech Republic, Judgment (13 November 2007, GC). This case was analysed in Chap. 2 (Sect. 2.​7.​2) on the right to education. It will be discussed further in Chap. 5 (Sect. 5.​2.​4) on the significance of international and European monitoring and litigation as a data source on Roma. Cahn (2004), p. 35.

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 40. Ramsay (2006), p. 2.

The notions equality and anti-discrimination were introduced in Chap. 2 (Sect. 2.​1).

For an overview of the main benefits of ethnic data collection for equality and anti-discrimination purposes, see Sect. 4.1. Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 40.

ERRC (2013), p. 42. This could be useful in legal proceedings or when reporting on the situation of human rights to international and European monitoring bodies, as explained in Sects. 4.1.4 and 4.1.5 on the benefits of ethnic data collection. The uncovering of discrimination and the identification of good practices was identified one of the main benefits of ethnic data collection in Sect. 4.1.2. See also Sect. 4.1.4 on the implementation, monitoring and evaluation of policies and Sect. 4.1.5 on the role of ethnic data in legal proceedings.

Alidadi (2017), p. 18.

Report of the High Commissioner for Human Rights containing a draft basic document on the development of a racial equality index (31 January 2006), para. 71. Makkonen (2010), pp. 227, 237 and 238. Gray (2009), pp. 56 and 57. Simon (2007), p. 16. Rallu et al. (2006), p. 535. Ramsay (2006), p. 2.

Farkas (2017), p. 14.

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 41.

Id. at para. 41.

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 23. Hermanin and Atanasova (2013). ENAR (2012), p. 7. Muigai (2000), p. 5. Gray (2009), pp. 56–58. Oppenheimer (2008). Simon (2007), pp. 14 and 16. Ringold et al. (2005), p. 64. Morning (2005), p. 41. As will be explained in Chap. 5 (Sect. 5.​6.​4), dissemination of data on Roma may not promote anti-Gypsyism.

Report of the Special Rapporteur on racism, racial discrimination, xenophobia and related forms of intolerance: Follow-up to and implementation of the Durban Declaration and Programme of Action (19 May 2009), para. 23. UNDP, Gender Aspects and Minority Data: An Illustrative Case of Roma Women in Southeast Europe (12 September 2006), para. 29.

Alidadi (2017), p. 18.

The way in which the data are reported on and disseminated, will also play an important role. This will be discussed in Chap. 5 (Sect. 5.​6.​4). The role of the media was underlined in Chap. 1 (Sect. 1.​2.​1) on the present-day situation of Roma in Europe. See Chap. 5 (Sect. 5.​8) on the importance of awareness-raising and active participation when collecting ethnic data on Roma. These key principles also play an important role in the framework of positive action, as will be considered generally in Chap. 6 (Sect. 6.​3.​2) and specifically in relation to the Roma minority in Europe in Chap. 9 (Sects. 9.​2.​1 and 9.​2.​4).

Report of the Special Rapporteur on racism, racial discrimination, xenophobia and related forms of intolerance: Follow-up to and implementation of the Durban Declaration and Programme of Action (19 May 2009), paras. 20 and 25. Eide (2013), pp. 49 and 50. Hermanin and Atanasova (2013). Ringelheim (2011), p. 1683. Ringelheim and De Schutter (2010), Foreword. Gray (2009), p. 58. Covrig (2004), p. 94.

Other examples of misuse of sensitive data include the Rwandan genocide of Tutsis (1994), Apartheid in South-Africa (1948–1991), forced migration of Japanese Americans in the USA (1941–1945) and of minorities in the Soviet Union (1948–1990s), and the forced removal of American Indians from their lands in the USA (nineteenth century). Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 42. Report of the Special Rapporteur on racism, racial discrimination, xenophobia and related forms of intolerance: Follow-up to and implementation of the Durban Declaration and Programme of Action (19 May 2009), paras. 20 and 25. Report of the High Commissioner for Human Rights containing a draft basic document on the development of a racial equality index (31 January 2006), paras. 72 and 73. Chopin et al. (2014), p. 59. Hermanin et al. (2013b), pp. 5 and 8. Wrench (2013), p. 80. Makkonen (2010), p. 238. Ringelheim (2008/9), p. 71. Oppenheimer (2008), p. 746. Simon (2007), pp. 14–16. Ringelheim (2006/7), p. 55. Seltzer and Anderson (2001), pp. 484, 486–488, 492, 493, 501 and 502.

ENAR (2012), p. 8. Makkonen (2010), p. 225. Jacobs et al. (2009), p. 68. Makkonen (2006), p. 13. Seltzer and Anderson (2001), pp. 482–496.

Some African countries still use ethnic categories to dominate certain groups. Rallu et al. (2006), p. 534.

ERRC (2013), p. 52.

Id.

The importance of awareness-raising and active and meaningful involvement of all stakeholders in ethnic data collection practices was addressed in Chap. 3 (Sect. 3.​7.​5).

For the reasons that will be explained in Chap. 5 (Sect. 5.​8), awareness-raising, active participation and genuine political will are identified as key principles in data collection practices on Roma.

See Chap. 3 (Sects. 3.​3–3.​7) for a discussion on these data protection rules and principles.

ECRI, General Policy Recommendation No. 11: Combating racism and racial discrimination in policing (29 June 2007), para. 1. FRA (2010c), pp. 5, 6, 10 to 13 and 15. FRA (2010b), p. 1. De Schutter and Ringelheim (2008), p. 363. Ethnic profiling was defined in Sect. 4.1.5. See also Chap. 3 (Sect. 3.​5.​3) for a brief introduction to sensitive data processing for profiling purposes in the police sector. Suggested further reading on how unlawful profiling can be prevented: FRA (2018).

Hermanin and Atanasova (2013). FRA (2010c), pp. 9, 10, 26 and 41. De Schutter and Ringelheim (2008), pp. 359, 360, 362 and 383. Office of the High Commissioner on National Minorities (2006), p. 23.

Report of the Special Rapporteur on contemporary forms of racism, racial discrimination, xenophobia and related intolerance on his visit to Spain (6 June 2013), para. 51. Report of the Special Rapporteur on contemporary forms of racism, racial discrimination, xenophobia and related intolerance on his mission to the United States of America (28 April 2009), paras. 101 and 102. FRA (2010c), p. 5.

Farkas (2017), p. 31. The Equality Data Initiative project, an initiative of the Open Society European Policy Institute (OSEPI), was implemented in 2013 and 2014 by OSEPI, Migration Policy Group and ENAR in Bulgaria, France, Germany, Hungary, Ireland, Romania and Sweden. For more information, see: https://​www.​migpolgroup.​com/​_​old/​anti-discrimination-equality/​equality-data-initiative/​ (Accessed 17 March 2019).

See Sect. 4.1.4 on the monitoring and evaluation of policies as one of the main benefits of ethnic data collection. FRA (2010c), p. 25. De Schutter and Ringelheim (2008), pp. 363, 376, 378 and 383. Makkonen (2006), p. 13. Simon (2007), pp. 20 and 68.

Office of the High Commissioner on National Minorities (2006), p. 24.

Commission Communication, Joint Report (17 January 2014), p. 6. Commission Communication, Application of Directive 2000/43/EC of 29 June 2000 implementing the principle of equal treatment between persons irrespective of racial or ethnic origin (30 October 2006), p. 9. Hermanin and Atanasova (2013). FRA (2010c), p. 25. Makkonen (2010), pp. 211 and 225. Muigai (2000), p. 3. Ringelheim (2008/9), p. 71. Makkonen (2006), pp. 13 and 85. Hollo (2006), pp. 24, 25, 51 and 52. Ramsay (2006), p. 3.

EU data protection is interpreted too restrictively in Austria, Belgium, Bulgaria, Croatia, the Czech Republic, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, the Netherlands, Poland, Portugal, Slovakia, Spain, Turkey. Lamberts et al. (2014), p. 11. This was previously addressed in Chap. 3 (Sects. 3.​3 and 3.​5) on the core data protection rules.

Alidadi (2017), p. 20. Ringelheim and De Schutter (2010).

See the analysis of the key rules and principles that apply to personal and sensitive data collection and processing in Chap. 3 (Sects. 3.​3–3.​7). The two main legal instruments are Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data (28 January 1981), including changes introduced by the Protocol amending Convention 108 (18 May 2018) (Convention 108+) and Regulation 2016/679 of the European Parliament and of the Council on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (27 April 2016) (GDPR). These instruments were first introduced in Chap. 2 (Sects. 2.​6.​2–2.​6.​3).

Alidadi (2017), p. 20.

Farkas (2017, p. 6) cites Denmark and Portugal as examples where this is the case.

See Chap. 5 (Sect. 5.​8.​3) on the key importance of political will to collect data on Roma.

End-of-mission statement on Romania, by Professor Philip Alston, United Nations Human Rights Council Special Rapporteur on extreme poverty and human rights (11 November 2015).

Gray (2009), pp. 62 and 63. Wrench (2007), pp. 2, 5, 79 and 83. Rose Tornstad and Otsby (2005), pp. 22 and 23. Wrench and Modood (2000), pp. 24 and 25.

Chopin et al. (2014), pp. 53–55.

Farkas (2017), p. 46. The author will come back to this point in Chap. 5 (Sect. 5.​6) when considering the analysis of data sets on Roma populations.

See Chap. 5 (Sect. 5.​6) on the factors impacting the analysis of data sets on Roma populations.

Wrench (2013), pp. 83 and 84. Makkonen (2010), p. 220. Makkonen (2006), pp. 14 and 44.

UNSD (2008b), paras. 1.4 to 1.7, 1.60 and 18. Ringelheim (2008/9), p. 63. Haug (2001), p. 304.

Information in population registers remains normative and is thus not anonymised. Vital statistics include data on births, deaths, marriages and divorces. Ringelheim (2008/9), p. 63. Makkonen (2006), pp. 14 and 44.

See, for instance, the EU Labour Force Survey. For other examples, see: FRA (2012), pp. 31 and 168. FRA (2011), pp. 16 and 21. FRA (2009a), pp. 40–42. FRA (2007), p. 54. Waldron (2011), p. 48. UNSD (2008a), p. 2. Makkonen (2006), p. 14.

The Scandinavian countries have developed a good system to connect various official statistics. For instance, Finnish registers and administrative data sources are used for statistical data purposes. Suggested further reading: Statistics Finland (2004). Waldron (2011), p. 48. Simon (2007), p. 27.

Makkonen (2010), p. 220.

Rallu et al. (2006), p. 536. Sect. 4.4 will look at ethnical classification.

Makkonen (2006), p. 44.

This is especially true in Austria, Slovenia and the Nordic countries where Censuses are conducted exclusively via registers that are updated annually. This also reduces the cost of the Census considerably. Many other countries only conduct the Census once every decade. Abdikeeva (2014), p. 10. Makkonen (2010), p. 220. Makkonen (2006), p. 43.

Wrench (2013), pp. 83 and 84. Makkonen (2010), p. 220. Makkonen (2006), pp. 43 and 44. Wrench (2007), pp. 5 and 6.

Makkonen (2006), p. 44. Ethnic monitoring will be considered as a data source in Sect. 4.3.3.

The advantages of official statistics as a data source in ethnic data collection practices were identified in Sect. 4.3.1.1.

Countries in North America, South America and Oceania are considerably more likely to inquire about ethnicity in the Census. The wide variety of terminology, categories and answer formats of ethnic affiliation question in Censuses will be considered in Sect. 4.4.3. Alidadi (2017), p. 25. Wrench (2013), p. 84. Morning (2005), pp. 1 and 14–22. Haug (2001), p. 304.

Surdu (2019, p. 13) is of the opinion that “(a)s for the censuses, it would be desirable to see a stronger European consensus for keeping ethnicity (and religion) a matter of one’s private life and not one of state knowledge about citizens”.

UNSD (2008b), para. 2.160. United Nations Economic Commission for Europe (UNECE) (2006), p. 95 and Appendix I.

Of the 23.565 interviewed minority respondents, 65% said to be in favour. EU-MIDIS provides baseline comparative data on experiences of discrimination of ethnic minorities and immigrants across Member States. A second survey (EU-MIDIS II) was conducted in 2016, but no similar question was asked. FRA (2010a), p. 17. FRA (2009b), pp. 27 and 272. The Special Eurobarometer 263 on Discrimination in the EU was conducted in 2006 and uncovered a similar willingness within the general population (75% of respondents) to provide anonymous information in the Census in order to improve the fight against ethnic discrimination (19% were opposed to doing so, while 6% did not know). See: European Commission (2007), p. 28.

Farkas (2017), p. 7.

As will be explained in Chap. 5, this poses a particular problem in relation to ethnic data collection on the Roma minority in Europe. See Sect. 5.​2.​1 on the limitations of official statistics in the Roma context and Sect. 5.​1.​3 on the multiple reasons for the widespread lack of (reliable) data on Roma. UNDP, Gender Aspects and Minority Data: An Illustrative Case of Roma Women in Southeast Europe (12 September 2006), para. 26. Haug (2001), p. 305.

This will be discussed further in Sect. 4.5.2.2 on the shortcomings of the self-identification approach. See also Chap. 5 on the multiple reasons for the widespread lack of reliable data on Roma (Sect. 5.​1.​3) and on the reluctance among Roma in Europe to self-identify in data collection practices (Sect. 5.​4.​1).

For instance, being refused entrance to a disco on the basis of one’s race or ethnic origin does not have similar consequences on the socio-economic status of a person as repeatedly being discrimination when applying for jobs. Farkas (2017), p. 7. Makkonen (2010), p. 220. Makkonen (2006), p. 43.

Farkas (2017), p. 22. Makkonen (2010), pp. 220 and 221. Makkonen (2006), p. 43. Wrench and Modood (2000), p. 25.

Alidadi (2017), p. 18. Makkonen (2010), p. 221. Ringelheim (2008/9), p. 63.

Abdikeeva (2014), p. 11. See Sects. 4.4 and 5.​5 for an overview of the different ethnical classification and ethnical identification methods.

Report of the High Commissioner for Human Rights containing a draft basic document on the development of a racial equality index (31 January 2006), paras. 50 to 53. UNDP (2010), p. 95.

The EctHR clearly states that indirect discrimination can also be proved without statistical evidence. EctHR, Oršuš and Others v. Croatia, Judgment (16 March 2010, GC), paras. 153 and 176. EctHR, Opuz v. Turkey, Judgment (9 June 2009), paras. 98, 99, 193 and 198. EctHR, D.H. and Others v. the Czech Republic, Judgment (13 November 2007), paras. 137, 164, 180, 187, 188 and 190 to 192. ECSR, European Roma Rights Centre v. Italy, Decision (7 December 2005), paras. 22 to 24. ECSR, European Roma Rights Centre v. Greece, Decision (8 December 2004), paras. 27, 28, 40 and 41.

The significance of international and European monitoring and litigation as a data source on Roma will be considered in Chap. 5 (Sect. 5.​2.​4).

Commentary on Effective Participation of Persons Belonging to National Minorities in Cultural, Social and Economic Life and in Public Affairs (27 February 2008), para. 30. Examples from ACFC country monitoring: Third Opinion on the Czech Republic (1 July 2011), paras. 36 and 38. Third Opinion on Hungary (18 March 2010), para. 58. See also: ECRI, General Policy Recommendation No. 4: National surveys on the experience and perception of discrimination and racism from the point of view of potential victims (6 March 1998), paras. 1, 6 and 9. Hollo (2006), p. 24.

Estonia produces new data by linking data sets. Farkas (2017), pp. 8 and 35.

In order to do so, it must first be mapped which data are available and can be used for equality and anti-discrimination purposes. Sweden has done this. Alidadi (2017), p. 27. Farkas (2017), pp. 8 and 23.

Alidadi (2017, p. 18) explains that data quality “is assessed based on its reliability, validity, scope and cost effectiveness”. The implementation, monitoring and evaluation of policies was cited as one of the benefits of ethnic data collection in Sect. 4.1.4. The link between ethnic data and positive action will also be highlighted in Chap. 6 (Sect. 6.​3.​4).

For examples of academic and ad hoc research, see: FRA (2009a), pp. 92 and 93. FRA (2008), pp. 36 and 37. FRA (2007), pp. 85–90, 127 and 128. Makkonen (2010), p. 223. Ringelheim (2008/9), pp. 65–67. Makkonen (2006), p. 14.

Farkas (2017), p. 27. Ringelheim (2008/9), pp. 64 and 65. Suggested further reading on the key role that equality bodies play in brokering change with regard to data collection, even if they are not always successful in their efforts: Farkas (2017), pp. 20 and 25.

Alidadi (2017, p. 26) cites the Netherlands as a good example.

Other terms used are situation testing, practice testing and matched-pair field experiments. This practice is quite common across the EU. See, for example: FRA (2013), pp. 192, 193 and 195. FRA (2012), pp. 165–167 and 171. Farkas (2011), pp. 47 and 48. FRA (2011), pp. 116 and 117. FRA (2009a), pp. 38, 39, 41 and 45. Rorive (2009), pp. 41–88. FRA (2008), pp. 10, 50, 58, 87, 108 and 116. FRA (2007), pp. 9, 33, 39, 40, 58, 59, 78, 79, 86, 150, 158 and 159. More on this: Wrench (2013), pp. 86 and 87. Bell (2007), p. 255. Makkonen (2006), p. 46. Fix and Turner (1998), p. 1.

Wrench (2013), pp. 86 and 87.

Discrimination testing is also used in France, which is traditionally hostile towards ethnic data. Makkonen (2010), p. 222. Ringelheim (2008/9), pp. 65–67. Makkonen (2006), p. 46.

Makkonen (2010), p. 221. Makkonen (2006), pp. 42 and 46. Wrench (2007), p. 9.

Surveys may be anonymous or non-anonymous. Makkonen (2010), p. 222. Ringelheim (2008/9), pp. 65–67. Makkonen (2006), p. 45.

Examples include: Alidadi (2017). Farkas (2017). European Commission (2008b). Makkonen (2006). Olli and Kofod Olsen (2006). Olli and Kofod Olsen (2005). Simon (2004), p. 27. Reuter et al. (2004).

Alidadi (2017), pp. 23 and 24. Farkas (2017), pp. 18 and 19.

Alidadi (2017), p. 24. Lamberts et al. (2014).

Wrench (2013), p. 85. Official statistics were discussed in Sect. 4.3.1.

Wrench (2013), p. 85.

Alidadi (2017), p. 18, 22 and 23. Makkonen (2010), 223. Simon (2005), pp. 13 and 14.

For instance, Eurobarometer surveys make it possible to gain insight into a variety of discrimination related topics, including perceptions and experiences of discrimination in the EU and the effectiveness of national anti-discrimination policies. Alidadi (2017), p. 24. Farkas (2017), p. 18.

FRANET replaced the European Racism and Xenophobia Network (RAXEN) and the network of academic and other legal experts (FRALEX) in July 2011. For more on RAXEN, see for example: European Monitoring Centre on Racism and Xenophobia (EUMC) (2005), pp. 7–13. Regulation 168/2007 of the Council establishing a European Union Agency for Fundamental Rights (15 February 2007), recital 14 and arts. 4.1(a)(c-g), 6.1(a) and 6.3.

Regulation 168/2007 of the Council establishing a European Union Agency for Fundamental Rights (15 February 2007), recitals 5 to 7, 10, 12, 27 and 30 and arts. 1, 4.1 (a)(b), 6 and 29.

Farkas (2017, pp. 33 and 34) includes several national examples of discrimination testing. Alidadi (2017), p. 18. Wrench (2013), p. 90. Makkonen (2006), p. 47. Hollo (2006), p. 48. Riach and Rich (2002), F513.

Wrench (2013), p. 85 and 86. Makkonen (2010), p. 223. Gray (2009), p. 59. Makkonen (2006), pp. 14, 42, 46 and 84.

Under- and over-reporting in relation to complaints data will be addressed in Sect. 4.3.4.2. Wrench (2013), p. 90.

Wrench (2013), p. 90.

For instance, employers stated in survey that they do not hire ethnic minorities, because their clients request them not to (Belgium), because they consider them to be too macho or lacking ambition (Germany), or because of their ethnicity (United Kingdom). Wrench (2013), pp. 85, 86 and 89. Makkonen (2010), p. 223. Gray (2009), p. 59. Makkonen (2006), pp. 14, 42, 46 and 84.

Makkonen (2010), p. 222. Makkonen (2006), p. 46.

Increased social acceptability of prejudices indicates an increased risk of discrimination. Makkonen (2010), p. 222. Makkonen (2006), p. 46.

For an example of a victim survey, see: EUMC (2006), pp. 48 and 49. Wrench (2013), p. 86. Makkonen (2010), p. 222. The limitations of complaints data as a data source will be addressed in Sect. 4.3.4.2.

Makkonen (2010), p. 223.

Farkas (2017), p. 8. Farkas (2017), pp. 27 and 28.

For instance, discrimination testing can not assess promotion in the workplace. Surveys could fill this gap. Alidadi (2017), p. 18. Wrench (2013), p. 86. Bond et al. (2010), p. 7. Makkonen (2010), p. 222. Makkonen (2006), p. 47.

France has a legal framework on discrimination testing since 2006. Alidadi (2017), p. 26.

Riach and Rich (2004), pp. 459, 462, 463 and 465. Riach and Rich (2002), pp. 482 and 484.

Bond et al. (2010), p. 6.

Haug (2001), p. 304.

Examples include the European Social Survey, the Household Budget Survey and the European Labour Force Survey. Wrench (2013), p. 84. Bond et al. (2010), p. 9. Ramsay (2006), p. 6.

See Chap. 5 (Sects. 5.​5.​1–5.​5.​3) on attainable ways to address difficulties in sampling Roma through surveys. Gray (2009), p. 57.

Ethnical classification will be discussed in Sect. 4.4. Wrench (2013), p. 84. Bond et al. (2010), p. 9. Ramsay (2006), p. 6.

Abdikeeva (2014), p. 11.

Alidadi (2017), pp. 23 and 24. Farkas (2017), pp. 16–18 and 40. Proxies for ethnicity will be discussed in Sect. 4.5.4. See also Chap. 5 (Sect. 5.​4.​2) on the use of proxies for Roma ethnicity.

The other groups that are overlooked include descendants of immigrants (from the second generation up), Afro-Europeans and Muslims. In order to collect reliable and comparable data in a cost-effective way, it is recommended that these surveys include more equality and anti-discrimination specific data. Alidadi (2017), pp. 23 and 24. Farkas (2017), pp. 16–18 and 47. Makkonen (2016), p. 11 and 60–62.

National research indicates that data collection on discrimination experiences is often received with reservations for this reason. Farkas (2017), p. 7. Wrench (2013), p. 89. Makkonen (2010), p. 222. Makkonen (2006), pp. 45 and 46. Wrench and Modood (2000), pp. 31 and 32.

Bond et al. (2010), pp. 5 and 6.

Wrench (2013), p. 89. As will be discussed in Chap. 5 (Sect. 5.​5.​4), the type of interview, the choice of interviewers and the questions asked have consequences for the quality of data collected on Roma.

Collection of data on gender, address, seniority and union membership is standard. Makkonen (2010), pp. 223 and 224. Makkonen (2006), p. 96.

Makkonen (2010), pp. 223 and 224.

Alidadi (2017), p. 17. ENAR (2015), p. 3. Makkonen (2010), p. 224. Ringelheim (2008/9), p. 64. Makkonen (2006), p. 96. Simon (2005), pp. 14 and 15.

ENAR (2012), p. 9. Makkonen (2010), p. 218. Makkonen (2006), p. 41. UK Commission for Racial Equality (2002), pp. 3 and 5.

UK Commission for Racial Equality (2002), p. 3. Makkonen (2010), p. 224.

RED, recital 23 and art. 11. The absence of an explicit legal obligation in EU law to collect ethnic data was cited in Chap. 3 (Sect. 3.​1.​1).

Member States must report regularly to the European Commission on national implementation of the RED. RED, arts. 14, 16 and 17. Simon (2005), pp. 13–16.

Examples include, among others, Belgium, Hungary and the UK. Makkonen (2006), p. 41.

Eurobarometer surveys monitor the evolution of public opinion in the EU on a variety of topics such as social situation, health and culture. European Commission (2012), p. 9. European Commission (2009), pp. 48–52. European Commission (2008a), pp. 30–33 and 84. European Commission (2007), p. 23.

Makkonen (2010), pp. 218, 223 and 224. Makkonen (2006), pp. 42 and 96.

United Kingdom (UK), Race Relations (Amendment) Act (2000). ERRC (2013), p. 52. McDonald and Negrin (2010), p. 13.

The public sector equality duty entered into force on 5 April 2011. The protected characteristics include age, disability, gender reassignment, marriage and civil partnership, pregnancy and maternity, race, including ethnic or national origin, color or nationality, religion or belief, sex and sexual orientation. UK, Equality Act (England and Wales) (April 2010), Part 11 Chap. 1 Sect. 149.

UK, Equality Act 2010 (Specific Duties) Regulations 2011 (England and Wales) (9 September 2011), regulations 2 and 3.

ENAR (2015), p. 11.

Makkonen (2010), pp. 218, 223 and 224. Makkonen (2006), pp. 42 and 96. Chap. 6 (Sect. 6.​3.​4) expands on the connection between ethnic data and positive action.

Lamberts et al. (2014), p. 11. Makkonen (2010), p. 224.

Alidadi (2017), p. 17. Makkonen (2010), p. 224. Ringelheim (2008/9), pp. 67 and 68. Official statistics were analysed as a data source in Sect. 4.3.1.

Barton (2004), pp. 24 and 25. Support of discrimination claims in legal proceedings was cited as the fifth benefit of ethnic data collection in Sect. 4.1.5.

ENAR (2012), p. 9. While specifically about governments’ fear about the exposure of bad policies, see Sect. 4.2.1.

Makkonen (2006), pp. 42 and 96.

Id. at pp. 98 and 236.

For examples of complaints data, see: FRA (2013), pp. 180–188 and 269. FRA (2012), pp. 156–163. FRA (2011), pp. 84–87 and 121–123. FRA (2009a), pp. 21–29 and 33. FRA (2008), pp. 27–36. FRA (2007), pp. 25, 26, 29, 30, 115–118, 121–126, 131 and 132. Makkonen (2010), p. 221.

As explained in Chap. 3 (Sect. 3.​1.​1), art. 13 RED obliges Member States to install equality bodies tasked with providing independent assistance to victims in discrimination cases, conducting independent discrimination surveys and publishing reports and recommendations on discrimination issues. All equality bodies collect complaints data and some try to improve their data collection through collaboration with NGOs. Farkas (2017), pp. 7, 8, 20, 25 and 28. Wrench (2013), p. 84. Makkonen (2010), p. 221.

Office of the High Commissioner on National Minorities (2006), p. 27.

Id.

Farkas (2017), p. 29.

Id. at pp. 7 and 24.

Farkas (2017, p. 24) mentions Austria and Denmark as examples of Member States where this is the case.

Wrench (2013), pp. 84 and 85. Wrench (2007), pp. 7 and 8. Makkonen (2006), p. 45.

Makkonen (2010), p. 221.

Commission Communication, Joint Report (17 January 2014), p. 6.

FRA (2014), p. 15. FRA (2011), pp. 83 and 84. FRA (2009a), pp. 21 and 23. FRA (2008), pp. 17, 21, 24, 48, 49, 97 and 115. FRA (2007), pp. 79–84. Wrench (2013), pp. 84 and 85.

Abdikeeva (2014), p. 14.

Research data were identified as a data source in Sect. 4.3.2.

See Chap. 5 (Sect. 5.​2.​3) on large gaps in complaints data on discrimination against Roma.

Farkas (2017), pp. 7, 8 and 24.

Id. at pp. 7 and 24.

Farkas (2017), p. 37. ENAR (2012), p. 8. Ringelheim (2011), p. 1683. Ringelheim (2006/7), p. 65. Suggested further reading: Nobles (2002), pp. 43–70. Ethnical identification and the use of proxies will be discussed in Sect. 4.5. The notions race and ethnicity were introduced in Chap. 2 (Sect. 2.​2). In Chap. 3 (Sect. 3.​2), it was considered how racial and ethnic origin can be defined for data collection purposes.

Waldron (2011), p. 48.

Section 4.5 will consider how people can be ethnically identified with one or multiple ethnic categories for data collection purposes.

A direct approach is one that involves data collection on discrimination grounds such as racial or ethnic origin, compared to an indirect approach which relies on proxies for ethnicity. Alidadi (2017), p. 24. Ethnical identification approaches will be analysed in Sect. 4.5.

Ringelheim (2011), pp. 1685 and 1686.

Makkonen (2006), p. 81. Haug (2001), pp. 305, 308 and 309. See Chap. 2 (Sect. 2.​2) for an introduction to the notions race and ethnicity. Chap. 3 (Sect. 3.​2) addressed how racial and ethnic origin could be defined for data collection purposes.

Simon (2007), p. 27. Olli and Kofod Olsen (2005).

CERD Committee, General Recommendation No. 24: Article 12 of the Convention Women and Health (27 August 1999), paras. 1 and 2. Human Rights Committee, General Comment No. 23: The rights of minorities (Art. 27) (8 April 1994), para. 5.2. ACFC, First Opinion on Poland (27 November 2003), para. 17.

The well-established principle of international human rights law that the existence of a minority is a question of fact, not law, was introduced in Chap. 3 (Sect. 3.​2.​2) on the context-dependency and variability of ethnicity.

Makkonen (2006), p. 78. Section 4.5 will focus on the different approaches to ethnic identification.

Makkonen (2006), p. 76.

This was previously highlighted in Chap. 1 (Sect. 1.​1.​3) on the lack of uniform status of Roma across Europe. It will be discussed further in Sect. 4.4.3 on the wide variety in terminology, categories and answer formats used.

In Chap. 1 (Sect. 1.​1.​3), when discussing the conditions that help determine the existence of a minority in a given situation, the presence of different ethnic, religious and/or linguistic characteristics among a numerically inferior group was highlighted as well as the motivation to preserve the culture, traditions, religion and language of the group. Objective and subjective elements are thus involved.

The impact of answer formats in ethnic data collection practices will be discussed in Sect. 4.4.3. See also Chap. 5 (Sect. 5.​3.​2) on the importance of including open-ended categories when collecting ethnic data on Roma in Europe.

See Sect. 4.4.3 on the wide variety of answer formats used to inquire about ethnicity.

As will be explained in Chap. 5 (Sect. 5.​3.​2), social scientists could help determine the feasibility and operationalibility of the available methodological choices.

The advantages and disadvantages of the four different ethnical identification methods will be reviewed in Sect. 4.5.

Open-ended answer formats will be discussed further in Sect. 4.4.3 on the wide variety in answer formats used. See also Chap. 5 (Sect. 5.​3.​2) on the importance of including open-ended categories when collecting data on Roma in Europe.

The applicable anti-discrimination, minority and data protection rules will inevitably influence the choice of ethnic classifications. Ringelheim (2008/9), pp. 92 and 127. Makkonen (2010), p. 236. Makkonen (2006), p. 78. Simon (2007), pp. 27 and 41. Haug (2001), pp. 303, 308 and 309.

Hermanin et al. (2013a), p. 139.

Krizsán (2011), pp. 1393 and 1394.

Makkonen (2010), p. 236. Makkonen (2006), p. 78.

Commission Green Paper, Equality and non-discrimination in an enlarged European Union (28 May 2004), pp. 5, 10 and 21. CERD Committee, General Recommendation No. 15: Measures to eradicate incitement to or acts of discrimination (Art. 4) (15 September 1993), paras. 2, 3 and 6. Makkonen (2002), p. 59. Chap. 1 (Sect. 1.​2.​3) expanded on intersectional discrimination faced by Roma.

Haug (2001), p. 308. The variability of ethnicity was highlighted in Chap. 3 (Sect. 3.​2.​2).

Ethnic questions in Censuses frequently change after categories become contested in a changing social, political or legal context. See, for instance, the ambiguous conceptualisation of the notion Roma in Chap. 1 (Sect. 1.​1.​2). Jacobs et al. (2009), p. 68. Ringelheim (2008/9), pp. 127 and 128. Makkonen (2006), p. 78. Simon (2004), pp. 34, 50, 54, 57 and 59. For an analysis of the evolution of racial categorisation in the American and Brazilian Census, see: Nobles (2002), pp. 49–66.

This follows from the subjective nature of the concept of ethnicity, as will be discussed in Sect. 4.5. For example, the Hungarian census allows minorities to choose multiple answers when asked about ethnic identity. Banton (2001, p. 67) stresses the importance of asking about ethnic origins rather than ethnic origin, as is done in the UK Race Relations Act (England and Wales), in order to allow for multiple affiliations. See: Waldron (2011), p. 48. Ringelheim (2008/9), p. 130. UNSD (2008b), para. 2.162. Simon (2007), pp. 28, 29, 55 and 61. Haug (2001), pp. 308 and 309. On the other hand, some countries only allow the selection of one ethnic origin in the Census. Farkas (2017, p. 16) mentions Estonia, Romania and Slovakia.

This relates to the subjective component of ethnic identities, as explained in Chap. 3 (Sect. 3.​2.​3) and which will be considered further in Sect. 4.5 on ethnical identification approaches. Gray (2009), p. 58. Ringelheim (2008/9), p. 130. Simon (2005), p. 21. Haug (2001), pp. 308 and 309.

See Chap. 5 (Sect. 5.​3) on challenges to the construction of ethnic categories for Roma.

See Chap. 5 (Sect. 5.​3.​2) on challenges to the construction of ethnic categories for Roma. Close co-operation with minority representatives is equally important, as will be highlighted in Sect. 4.4.3.

Farkas (2017), p. 38.

These are typical examples of traditional immigration societies interested in ethnic data relating to the people’s immigration background or their ethnic ancestry. Rallu et al. (2006), pp. 537–541. Haug (2001), p. 308.

The current categories are the result of years of experimenting with several schemes based on geographic origin and combining race and ethnicity. Makkonen (2006), p. 74. Simon (2005), p. 20.

These are White (Irish; Gypsy or Irish Traveller; Other White), Mixed/multiple ethnic groups (White and Black Caribbean, White and Asian; White and Black African; Other Mixed); Asian/Asian British (Indian; Pakistani; Bangladeshi; Chinese; Other Asian); Black/African/Caribbean/Black British (African; Caribbean; Other Black); Other ethnic group (Arab; Any other ethnic group). The lack of transferability of the UK and Irish categories for Roma will be considered in Chap. 5 (Sect. 5.​3.​1) on the challenges to constructing ethnic categories for Roma.

From 1971 onwards, the UK Census inquired about the nationality and country of birth of the parents. This changed in 1991 because the information collected was considered to be unsatisfactory for equality and anti-discrimination purposes. Research between 1975 and 1989 shows that there was little opposition among minorities to include questions on ethnicity in the Census, but that there was concern about the way the questions would be formulated. According to Simon (2005, pp. 19 and 20; 2004, pp. 50 and 51), the ethnicity question had become commonplace in British statistics by the beginning of the twenty-first Century, with the promotion of equality as a determining factor in that process. UK Office for National Statistics (2012), pp. 2 and 12. Ringelheim (2008/9), pp. 107–110. Simon (2007), p. 61. Rallu et al. (2006), p. 539.

Report of the High Commissioner for Human Rights containing a draft basic document on the development of a racial equality index (31 January 2006), para. 17. Oppenheimer (2008), p. 736. Ringelheim (2008/9), pp. 102–105. Simon (2005), p. 20. Simon (2004), pp. 48 and 50.

Makkonen (2006), p. 75. The connection and overlap between race and ethnicity was highlighted in Chap. 2 (Sect. 2.​2.​2.​3) and Chap. 3 (Sect. 3.​2.​3).

The categories include White; Black or African American; American Indian and Alaska Native; Asian; Native Hawaiian and Other Pacific Islander.

As explained in Chap. 2 (Sect. 2.​2 on the notions race and ethnicity) and Chap. 3 (Sect. 3.​2 on defining racial and ethnic origin for data collection purposes), ethnic affiliation relates to linguistic and cultural criteria. Hispanics and Latinos can be of any race. Note that the category Hispanic was only introduced in 1970 after lobbying efforts from the Hispanic community. Report of the High Commissioner for Human Rights containing a draft basic document on the development of a racial equality index (31 January 2006), para. 18. Simon (2005), p. 20. Haug (2001), p. 308.

Simon et al. (2015), p. 11.

Haug (2001), p. 308. Part II of the book focuses on positive action. The connection between ethnic data collection and positive action was briefly explained in Chap. 1 (Sect. 1.​3) when introducing the two main topics of this book. It will be explored further in Chap. 4 (Sect. 4.1.4) on the implementation, monitoring and evaluation of positive action as one of the main benefits of ethnic data collection for equality and anti-discrimination purposes, and in Chap. 6 (Sect. 6.​3.​4) on the role of ethnically disaggregated data in positive action schemes.

Simon et al. (2015), pp. 11 and 12.

Makkonen (2006), pp. 73 and 78. Simon (2004), pp. 34, 50, 54, 57 and 59.

The 2000 round of Censuses took place from 1995 to 2004 in 147 countries. The UNSD keeps an online database of census questions on, among others, national and/or ethnic groups: https://​unstats.​un.​org/​unsd/​Demographic/​sconcerns/​popchar/​popcharMeta.​aspx (Accessed 28 January 2019). Dahal et al. (2007), pp. 4, 5 and 13. UNSD (2003), pp. 3–5 and 7. Morning (2005), pp. 1 and 14–22. Official statistics, including Censuses, were analysed in Sect. 4.3.1.

The term ethnicity is mostly used. Additionally, countries mostly use nationality (Eastern European countries and Asian countries from the Former Soviet Union), indigenous (North America, South America, Oceania and Africa) and race (North America). Morning (2015), p. 17. Morning (2005), pp. 1 and 14–22.

Farkas (2017), p. 36. Simon (2007), p. 41. Haug (2001), pp. 308 and 309. See also Chap. 5 (Sect. 5.​5.​4), where the importance of the quality of the questions asked will be highlighted.

Banton (2001), p. 67.

See, for instance, Germany, France and Sweden. Gerards (2007), p. 48. Ethnical identification methods will be considered in Sect. 4.5.

Rallu et al. (2006), p. 541. The notions race and ethnicity were introduced in Chap. 2 (Sect. 2.​2). See also Chap. 3 (Sect. 3.​2) on how to define racial and ethnic origin for data collection purposes.

It concerns the 1991 Census used in Great Britain (England, Wales and Scotland). Rallu et al. (2006), p. 541. The categories used in the UK and the US Census were cited in Sect. 4.4.2.

Jacobs et al. (2009), p. 68. Morning (2015), p. 17. Banton (2001), p. 66. The context-dependent character of racial and ethnic origin was addressed in Chap. 3 (Sect. 3.​2).

While the borders between concepts ethnicity, race and nationality may be fluid, they all refer to ancestry while each relies on a different manifestation of the shared roots: ethnicity looks at cultural practices or beliefs, race at physical traits, and nationality at geographic location. Simon (2007), pp. 18, 26 and 27. Morning (2005), pp. 3–5.

In Central and Eastern Europe, minority denotes a lesser status than nationality. Gray (2009), p. 58. See Chap. 1 (Sect. 1.​1.​3) for an exploration of the notion minority.

Recommendation 1735 of the Parliamentary Assembly on The Concept of “Nation” (26 January 2006), paras. 2 to 6.

UNSD (2008b), para. 2.162. UNSD (2014), p. 171. UNECE (2006), para. 428. UNSD (2003), p. 11.

Rallu et al. (2006), p. 542.

ENAR (2012), p. 8. Jacobs et al. (2009), p. 68.

By redefining and regrouping categories, certain groups could be suppressed or eliminated. Ramsay (2006), pp. 4 and 5. Haug (2001), p. 305.

This is the case in France, Greece (only recognises the Muslim minority) and Turkey to ensure unity and homogeneity. Minorities are then covered by the category others, where they are overlooked. Ramsay (2006), pp. 4 and 5. Haug (2001), p. 305. The lack of uniform status of Roma across Europe was addressed in Chap. 1 (Sect. 1.​1.​3). It should be noted that, as will be explained in Chap. 5 (Sect. 5.​3.​2) on ethnic data collection on Roma, lack of formal recognition of a minority does not necessarily hinder official data collection on that group, while formal recognition does not guarantee inclusion in official data collection practices.

Report of the Special Rapporteur on contemporary forms of racism, racial discrimination, xenophobia and related intolerance (19 August 2013), para. 71. CRC Committee General Comment No. 11: Indigenous children and their rights under the Convention (12 February 2009), para. 80. Report of the Independent Expert on Minority Issues on the Implementation of General Assembly Resolution 60/251 of 15 March 2006 entitled “Human Rights Council” (2 February 2007), para. 77. Durban Plan of Action (8 September 2001), para. 92(b). UNSD (2014), p. 170.

Farkas (2017), pp. 12, 38 and 41–43. Some good practices of minority involvement can be found though, as will be explained in Chap. 5 (Sect. 5.​3) on the ethnical classification of Roma identities for data collection purposes. See, more generally, Chap. 5 (Sect. 5.​8.​2) on the need for active participation of Roma in data collection practices.

Farkas (2017), p. 45.

Simon (2007), p. 41. Haug (2001), pp. 308 and 309.

Most South American and African countries rely on closed-end answers, most Asian countries on open-ended answers, and European countries on a mixed approach. Within the EU, some Member States use a precoded list (examples include Bulgaria, Cyprus, Hungary, Latvia, Lithuania and the UK), whereas others use an open-ended question format (for instance, Poland and Romania). Some countries take a mixed approach. Simon (2007), p. 41. Morning (2005), pp. 1 and 14–22.

Rallu et al. (2006), p. 542.

Id.

They acknowledge that an open question may result in a larger number of ethnic groups, which may cause coding and grouping problems to present the results. Rallu et al. (2006), pp. 541–543. Ethnical categorisation was discussed in Sect. 4.4.

UNSD (2008b), para. 2.162. The Roma minority is a very heterogeneous minority and this has consequences for the ethnical categorisation process, as will be explained in Chap. 5 (Sect. 5.​3.​1) on the complex and multiple nature of Roma identities.

Simon (2007), p. 41.

Haug (2001), pp. 308 and 309.

Moreover, the choice as to whether or not to disclose one’s ethnicity may not result into disadvantages.

See, for example: ACFC, First Opinion on Azerbaijan (22 May 2003), para. 21. ACFC, First Opinion on Ukraine (1 March 2002), para. 22. ACFC, First Opinion on Poland (27 November 2003), para. 24.

ECRI, General Policy Recommendation No. 14: Combating racism and racial discrimination in employment (22 June 2012), para. 10(a). ECRI General Policy Recommendation No. 13: Combating Anti-Gypsyism and Discrimination against Roma (24 June 2011) para. 14. Examples from ECRI country monitoring: ECRI, Fifth Report on Belgium (4 December 2013), para. 122. ECRI, Fourth Report on Finland (21 March 2013), para. 186. ECRI, Fourth Report on France (29 April 2010), para. 153. ECRI, Fourth Report on the Czech Republic (2 April 2009), para. 177.

Different methodologies not only limit data comparability at the national level, but also at EU level. The United Nations Principles and Recommendations for Population and Housing Censuses: Results of the Survey on Proposed Changes for the 2020 Census Round prepared by the UNSD (October 2013), paras. 75 to 77. Report of the Special Rapporteur on racism, racial discrimination, xenophobia and related forms of intolerance: Follow-up to and implementation of the Durban Declaration and Programme of Action (19 May 2009), paras. 9 and 27. Report of the Independent Expert on Minority Issues on the Implementation of General Assembly Resolution 60/251 of 15 March 2006 entitled “Human Rights Council” (2 February 2007), para. 72. Farkas (2017), pp. 16 and 37. UNSD (2010), p. 107. UNSD (2008b), paras. 2.160 and 2.162. UNECE (2006), para. 429. UNDP (2005), pp. 10, 17, 30, 36 and 37. UNSD (2003), pp. 2 and 10. National differences and limited data comparability were previously highlighted in Sect. 4.4.2 on ethnic classification as a complex, context-dependent and evolving process.

See, for instance: The United Nations Principles and Recommendations for Population and Housing Censuses: Results of the Survey on Proposed Changes for the 2020 Census Round prepared by the UNSD (October 2013), para. 76. UNSD (2014), p. 3. UNSD (2008b). Farkas (2017, pp. 7, 21 and 36) reports that the progressive recommendations made by UN statisticians in order to promote ethnic data collection have not yet been taken on board by regional and national stakeholders. As a results, only a few types of data collection methods (e.g. the UK) comply to a considerable extent to the UNSD’s recommendations. Methodological inconsistencies thus remain.

The EU statistical office Eurostat plays an important role in this regard. Regulation 1201/2009 of the Commission implementing Regulation (EC) No 763/2008 of the European Parliament and of the Council on population and housing censuses as regards the technical specifications on the topics and of their breakdowns (30 November 2009). Regulation 763/2008 of the European Parliament and of the Council on population and housing censuses (9 July 2008). Abdikeeva (2014), p. 10.

Hermanin et al. (2013a), pp. 138 and 139.

Alidadi (2017), p. 25. This will be addressed further in relation to the Roma minority in Chap. 5 (Sect. 5.​3.​1) on the influence of the complex and fluid nature of Roma identities on ethnical classification.

The United Nations Principles and Recommendations for Population and Housing Censuses: Results of the Survey on Proposed Changes for the 2020 Census Round prepared by the UNSD (October 2013), para. 76. Report of the Special Rapporteur on racism, racial discrimination, xenophobia and related forms of intolerance: Follow-up to and implementation of the Durban Declaration and Programme of Action (19 May 2009), paras. 9 (b) and (c). UNSD (2008b), para. 2.162. UNSD (2014), p. 170. UNECE (2006), para. 429.

This was previously explained in Chap. 3 (Sect. 3.​2.​3) when considering how to define racial and ethnic origin for data collection purposes. See also the discussion on the notions race and ethnicity in Chap. 2 (Sect. 2.​2).

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 55. The involvement of objective and subjective criteria in the determination of ethnic origin was mentioned in Sect. 4.4.1 and in Chap. 3 (Sect. 3.​2.​3).

Id.

Ringelheim (2013), p. 551. Ringelheim (2011), p. 1688. Ringelheim (2006/7), p. 67. Simon (2004), pp. 37–41.

This approach is sometimes also referred to as self-reporting. This approach is used in the UK. Ringelheim (2006/7), p. 67. Simon (2005), pp. 20 and 21.

The Netherlands and Sweden are two examples of countries collecting data on the basis of objective indicators such as birthplace. Chopin et al. (2014), pp. 4 and 7. Ringelheim (2006/7), p. 67. Simon (2005), p. 22. Objective criteria were previously discussed in Chap. 2 (Sect. 2.​2.​2.​2) on ethnicity as a social construct and in Chap. 3 (Sect. 3.​2) when considering how to define racial and ethnic origin for data collection purposes.

Ringelheim (2008/9), p. 135. Ringelheim (2006/7), p. 67. Simon (2005), pp. 21 and 22.

This is done in the USA to register the tribe affiliation of American Indians and in Australia to confirm Aboriginal status. Ringelheim (2011), p. 1686. Ringelheim (2006/7), p. 67. Makkonen (2006), p. 73. Simon (2005), p. 22. Simon (2004), pp. 34–37.

Simon et al. (2015), p. 4. Chopin et al. (2014), pp. 56–58. FRA (2007), p. 151. Cardinale (2007), p. 38. Ringelheim (2006/7), pp. 55 and 56.

Ringelheim (2013), pp. 50 and 51. Ringelheim (2011), p. 1686. Simon (2005), p. 20.

Declaration on the Rights of Persons Belonging to National or Ethnic, Religious and Linguistic Minorities (3 February 1992), art. 3.

CERD Committee, General Recommendation No. 8: Identification with a particular racial or ethnic group (Art. 1, paras. 1 & 4) (22 August 1990), para. 79.

CESCR Committee, General Comment No. 20: Non-Discrimination in Economic, Social and Cultural Rights (Art. 2, para. 2) (2 July 2009), para. 16.

UNDP (2010), p. 95.

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 54. Report of the Special Rapporteur on contemporary forms of racism, racial discrimination, xenophobia and related intolerance (19 August 2013), para. 82. Report of the Special Rapporteur on racism, racial discrimination, xenophobia and related forms of intolerance: Follow-up to and implementation of the Durban Declaration and Programme of Action (19 May 2009), paras. 9, 25, 26, 52 and 72. Muigai (2000), p. 5.

See, for example: ACFC, Third Opinion on Hungary (18 March 2010), paras. 19, 38 and 62. Framework Convention for the Protection of National Minorities (1 February 1995) (FCNM), art. 3. Explanatory Report to the FCNM (1 February 1995), paras. 35 to 34.

See, for example: ECRI, Fourth Report on Latvia (9 December 2011) CRI(2012)3, paras. 53, 54, 182 and 184. ECRI, General Policy Recommendation No. 11: Combating racism and racial discrimination in policing (29 June 2007), para. 41.

In the UK and Ireland, this practice is applied beyond the Census and also covers ethnic monitoring at public or private organisations. Ringelheim (2006/7), p. 67. Kertzer and Arel (2002), p. 34. The UN strongly encourages the use of self-identification in Censuses. See, for instance: UNSD (2008b), para. 2.162. Varieties in question format were addressed in Sects. 4.4.1 and 4.4.3. Censuses and their role in ethnic data collection was described in Sect. 4.3.1 on official statistics as a data source.

Farkas (2017), p. 7.

Id.

Other essential elements of one’s private life include, among others, name, gender, religion and sexual orientation. EctHR, Ciubotaru v. Moldova, Judgment (27 April 2010), paras. 71 to 74. EctHR, Connors v. the United Kingdom, Judgment (27 May 2004), para. 82. EctHR, Odièvre v. France, Judgment (13 February 2003), para. 29. EctHR, Christine Goodwin v. the United Kingdom, Judgment (11 July 2002, GC), para. 90. EctHR, Mikulić v. Croatia, Judgment (7 February 2002), para. 53. EctHR, Bensaid v. the United Kingdom, Judgment (6 February 2001), para. 47. EctHR, Chapman v. the United Kingdom, Judgment (18 January 2001, GC), para. 96. The notion privacy was analysed in Chap. 2 (Sect. 2.​5).

Ringelheim (2013), pp. 47, 48, 52, 53, 58 and 59. Waldron (2011), p. 48. Krizsán (2011), p. 1396. Ringelheim and De Schutter (2010), pp. 94–99. Cardinale (2007), pp. 37 and 38. Ringelheim (2006/7), pp. 53, 54, 66–69 and 77. Ramsay (2006), p. 2. Gachet (2001), p. 54. As explained in Chap. 3 (Sect. 3.​2.​3), the determination of ethnic origin for data collection purposes involves both objective and subjective criteria. Also, the focus on individual self-determination within the framework of the right to private life was highlighted in Chap. 2 (Sect. 2.​5.​2).

Examples from ACFC country monitoring: ACFC, Third Opinion on the Russian Federation (24 November 2011), paras. 32, 33, 36 and 37. ACFC, Third Opinion on Hungary (18 March 2010), paras. 19, 38 and 62. ACFC, Second Opinion on the United Kingdom (6 June 2007), para. 48. The wide variety in terminology, categories and answer formats used in ethnic data collection practices was addressed in Sect. 4.4.3.

Explanatory Report to the FCNM (1 February 1995), para. 35.

Considering that Moldova used to be part of the Soviet Union, the EctHR considered it to be a disproportional burden for the claimant to provide proof of the official records of his parents that were drafted by Soviet authorities. EctHR, Ciubotaru v. Moldova, Judgment (27 April 2010), paras. 57 and 58. The inclusion of positive obligations for States when protecting personal data through the right to private life was discussed in Chap. 3 (Sect. 3.​6.​2).

Mijović refers to the Sejdić and Finci judgment from 2009 where one had to declare affiliation with a constituent people in order to be eligible to stand for election. Refusal to do so resulted in exclusion from elections. The rule was introduced to ensure peace in a specific post-conflict situation. See: EctHR, Sejdić and Finci v. Bosnia and Herzegovina, Judgment (22 December 2009, GC), paras. 43 and 45.

Makkonen (2006), p. 85. For more on data security, see the analysis of the personal data protection rules in Chap. 3 (Sects. 3.​3–3.​7).

Makkonen (2010), p. 236. Makkonen (2006), p. 85.

This happened during the 1991 Census in the Former Yugoslav Republic of Macedonia (renamed the Republic of North Macedonia in February 2019) and the 2001 Census in Croatia. Makkonen (2010), pp. 236 and 237. Ringelheim (2008/9), p. 129. Makkonen (2006), pp. 75, 77 and 81. Ramsay (2006), pp. 2 and 3. Krizsán (2001), pp. 173–175, 193 and 194.

Ringelheim (2013), pp. 53 and 54. Ringelheim (2011), pp. 1688–1693. Ringelheim (2006/7), p. 69. Ramsay (2006), pp. 2 and 3. Simon (2005), p. 21. The fear of misuse of sensitive data to discrimination against ethnic groups was cited as the third risk or fear surrounding ethnic data collection in Sect. 4.2.3. The subjective element of ethnicity was mentioned in Chap. 3 (Sect. 3.​2.​3).

Report of the Special Rapporteur on combating racism, racial discrimination, xenophobia and related intolerance on the comprehensive implementation of and follow-up to the Durban Declaration and Programme of Action (20 August 2015), para. 58.

Alidadi (2017), p. 22. UNSD (2008b), para. 2.160. UNSD (2014), p. 171. This will be addressed further in Chap. 5 (Sect. 5.​8.​1) on the need for awareness-raising.

Alidadi (2017), p. 22. The European data protection framework was analysed in Chap. 3 (Sects. 3.​3–3.​7).

Makkonen (2010), pp. 236 and 237. Makkonen (2006), p. 77. Ringelheim (2006/7), pp. 70 and 71. Ramsay (2006), pp. 2 and 3. Haug (2001), p. 306. Krizsán (2001), p. 175. Part II of the book focuses on positive action.

See Chap. 5 on the reluctance among Roma to self-identify (Sect. 5.​4.​1) for under-reporting of Roma ethnicity and on fraudulous reporting of Roma identity by non-Roma in order to access special rights (Sect. 5.​4.​4) for over-reporting.

Ringelheim and De Schutter (2010), Foreword. Ringelheim (2006/7), pp. 66 and 67.

Ringelheim (2013), pp. 53, 54 and 57–59. Ringelheim (2011), pp. 1689 and 1692–1694. Ramsay (2006), pp. 2 and 3. Haug (2001), p. 309. The uncovering of discrimination was cited as the second benefit of ethnic data collection in Sect. 4.1.2, and the implementation, monitoring and evaluation of policies as the fourth one in Sect. 4.1.4. The link between positive action and ethnic data will also be emphasised in Chap. 6 (Sect. 6.​3.​4).

ACFC, Second Opinion on Slovakia (26 May 2005), para. 30.

ACFC, Third Opinion on Hungary (18 March 2010), paras. 42 and 43.

Ringelheim (2013, pp. 53 and 54; 2011, pp. 1692–1694) explains that while minority protection involves both objective and subjective components (see Chap. 1 Sect. 1.​1.​3, Chap. 3 Sect. 3.​2.​3 and Chap. 4 Sect. 4.4.1), ethnic discrimination goes beyond the FCNM and involves additional instruments such as the International Convention on the Elimination of All Forms of Racial Discrimination that apply more broadly than only to minorities (see Chap. 2 on equality and anti-discrimination in Sect. 2.​1 and on minority rights protection in Sect. 2.​4) and where the subjective component is less important due to the importance given to certain protected characteristics like ethnic, racial or national origin and how people are perceived by others. Chopin et al. (2014), p. 57. Simon (2005), p. 21. Farkas (2004), pp. 19–23.

Other groups include Afro-Europeans and European Muslims. Farkas (2017), p. 6.

Makkonen (2006), p. 98. Ringelheim (2006/7), pp. 54, 66–67 and 77.

CERD Committee, General Recommendation No. 8: Identification with a particular racial or ethnic group (Art. 1, paras. 1 & 4) (22 August 1990), para. 79.

Makkonen (2010), pp. 233 and 234. Makkonen (2006), p. 77.

Ringelheim (2006/7), p. 76.

UK, Fair Employment and Treatment (Northern Ireland) Order (1998), arts. 52 to 54. ACFC, Second Opinion on the United Kingdom (6 June 2007), paras. 47 to 49.

For more on positive action and the link with ethnic data, see Chap. 6 (Sect. 6.​3.​4).

The limits of identification on the basis of objective criteria will be discussed in Sect. 4.5.4.

UNDP, Gender Aspects and Minority Data: An Illustrative Case of Roma Women in Southeast Europe (12 September 2006), para. 33. Chopin et al. (2014), p. 57. Ringelheim (2008/9), pp. 135 and 136. Ramsay (2006), p. 3. Simon (2005), p. 21.

Makkonen (2010), pp. 237 and 238. Simon (2005), p. 22.

Id.

Rughiniș (2010), p. 356. See Chap. 5 (Sect. 5.​4.​4) for a discussion on this in relation to data collection on Roma.

Over-reporting was described as one of the possible shortcomings of the self-identification approach Sect. 4.5.2.2. This will be addressed further in Chap. 5 on ethnical identification by other Roma (Sect. 5.​4.​4) and when explaining how more accurate on Roma results may be obtained by combining different ethnical identification approaches (Sect. 5.​4.​5).

Individual self-determination within the framework of ethnical identification was addressed in Sect. 4.5.2.1 on self-identification as the preferred approach. See also Chap. 2 (Sect. 2.​5.​2) on the notion privacy and how it has been interpreted by international and European bodies.

See Sect. 4.5.2 on the importance of respecting personal autonomy of the data subject. The United Nations Principles and Recommendations for Population and Housing Censuses: Results of the Survey on Proposed Changes for the 2020 Census Round prepared by the UNSD (October 2013), para. 76. Ringelheim (2011), pp. 1693 and 1694. Ringelheim (2006/7), pp. 76 and 77.

Convention 108(+), art. 9. GDPR, arts. 12 to 22. Recommendation CM/Rec(97)18 of the Committee of Ministers to Member States concerning the Protection of Personal Data Collected and Processed for Statistical Purposes (30 September 1997), para. 7. For the Northern-Ireland case, see: ACFC, Second Opinion on the United Kingdom (6 June 2007), para. 47. For the Dutch case, see: The Netherlands, Wet Bescherming Persoonsgegevens (6 July 2000), art. 18. For more on the information requirement and data subjects’ right to access, see Chap. 3 (Sects. 3.​4.​2 and 3.​4.​3) on the fairness and the transparency of personal data processing.

See Chap. 5 (Sect. 5.​4.​5) on the combination of different identification approaches.

ENAR (2015), p. 7. Chopin et al. (2014), pp. 4, 7 and 48. Farkas (2007), p. 36. Ringelheim (2006/7), p. 67. Simon (2005), p. 22. Haug (2001), p. 308.

The proxies used vary among Member States. While some are considered sensitive, proxies are not always used for equality and anti-discrimination purposes. Chopin et al. (2014), p. 48. Makkonen (2010), p. 227. Möschel (2009), pp. 197–217.

Chopin et al. (2014), p. 48. Official statistics were mentioned as the first data source in Sect. 4.3.1.

Gray (2009), p. 62. Ringelheim (2006/7), pp. 54–56. Simon (2007), pp. 42, 47, 62 and 69.

ENAR (2012), p. 6. This was previously mentioned in Sect. 4.3.2.2 on the disadvantages of research data as a data source.

WHO (2010), p. 1. The limitations of the self-identification approach were cited in Sect. 4.5.2.

Haug (2001), p. 308. See Chap. 3 (Sect. 3.​2.​3) for a discussion on the inclusion of objective and subjective criteria in the determination of ethnic origin.

See Sect. 4.4.2 on ethnical classification as a complex, context-dependent and evolving process.

Ringelheim (2006/7), pp. 74 and 75.

Such ethnic monitoring is voluntary since 1 January 2004. The Dutch Minority Policy uses a specific definition of ethnic minority (allochtoon) and specifically targets (1) immigrants and their descendants for whom the authorities feel a special responsibility because they come from former Dutch colonies (Surinamese, Antillans, Arubans and Moluqans) or because they were recruited in the past to work in the Netherlands (Moroccans and Turks), who are in a structurally disadvantaged socio-economic situation. Such classification reflects those most vulnerable to and affected by ethnic discrimination in the Netherlands and can therefore be considered to be reasonably adapted to the specific Dutch context. The Netherlands, Wet Bescherming Persoonsgegevens (6 July 2000), arts. 18 and 23.1(a). The Netherlands, Wet Stimulering Evenredige Arbeidsdeelname Allochtonen (Wet SAMEN) (23 April 1998). Ringelheim (2006/7), pp. 74 and 75. Suggested further reading: Guiraudon et al. (2005), pp. 76–79. Jacobs and Rea (2005), pp. 37–40.

See, for example: ACFC, Second Opinion on Germany (1 March 2006), para. 26. ACFC, Second Opinion on Norway (5 October 2006), para. 31. The increasing contestation of the nationality requirement for national minorities was underlined in Chap. 1 (Sect. 1.​1.​3). See also: Henrard (2007), p. 12.

See, for instance: Organisation for Economic Co-operation and Development (2013), pp. 191–230. Pew Research Centre (2004). Schnapper et al. (2003), p. 38.

WHO (2010), p. 1.

Farkas (2017), p. 40.

This was previously highlighted in Sect. 4.3.2.2 when addressing the shortcomings of EU-wide surveys that collect data on citizenship rather than asking about one’s racial or ethnic origin. Proxies for Roma ethnicity will be reviewed in Chap. 5 (Sect. 5.​4.​2).

Id.

The lack of uniform status of Roma across Europe was addressed in Chap. 1 (Sect. 1.​1.​3).

This risk was previously mentioned in Chap. 1 (Sect. 1.​1.​3) when discussing the lack of uniform status among Roma in Europe.

WHO (2010), p. 1.

Farkas (2017), p. 46.

Farkas (2017), p. 47. The importance of consultations with and active participation of minorities in the construction of ethnic categories for Roma and, more broadly, in all stages of data collection will be highlighted in Chap. 5 (Sects. 5.​3 and 5.​8.​2 respectively). It was previously cited in Chap. 3 (Sect. 3.​7.​5) as the fifth organisational principle of ethnic data collection.

This the case in many Central en Eastern-European countries. Haug (2001), p. 308.

Haug (2001), p. 308.

See Chap. 5 (Sect. 5.42) on how proxies for Roma ethnicity produce insufficient data for equality and anti-discrimination purposes.

UNSD (2008b), paras. 2.162 and 3.71. UNECE (2006), para. 424.

Chopin et al. (2014), pp. 7, 49 and 58.

Id. at pp. 7 and 49. This was underlined in Chap. 3 when analysing the data protection rules and principles (Sects. 3.​3–3.​7).

Krizsán (2011), p. 1397. The notion social inclusion was introduced in Chap. 2 (Sect. 2.​3).

ENAR (2012), p. 11.

Id.

ENAR (2012), p. 11. Krizsán (2011), p. 1404.

Intersectional discrimination faced by Roma was described in Chap. 1 (Sect. 1.​2.​3). See also the reflections on the situation of Roma in Europe, particularly so the part on anti-Gypsyism, in Chap. 1 (Sect. 1.​2).

The importance of the context was cited in Sect. 4.5.3 on alternative or supplementary approaches to ethnical identification.

See Chap. 5 (Sect. 5.​4.​2) on the use of proxies to collect data on Roma.

Chap. 5 (Sect. 5.​4.​5) will expand on how combining different ethnical identification approaches could result in the collection of more accurate data on Roma.