Gender Aspects and Minority Data: An Illustrative Case of Roma Women in Southeast Europe

1. Gender Aspects and Minority Data: An Illustrative Case of Roma Women in Southeast Europe United Nations Development Programme Nadja Dolata and Susanne Milcher, Bratislava Regional Centre Group of Experts on gender statistics, UNECE, Geneva, 12 September 2006

2. Overview of presentation • Gender and minorities: the case for social inclusion and policy making • Conceptual framework • The Vulnerable Groups Survey, UNDP • The case of Roma women • Challenges to data collection • Data collection instruments • Recommendations

3. Gender and minorities: the case for social inclusion and policy making • The need for adequate data for policy formulation and monitoring. • To identify the causal relationships related to social exclusion. • Vulnerable groups exist in all societies and need to be included. • But how can minorities and gender aspects be included in policies and monitoring without adequate data to map the situation of these sub-populations?

4. Gender and minorities -Conceptual framework • Multiple stratifiers • Forms and ways of coping with inequality will always be determined by multiple stratifiers (ethnicity, class, disabilty, age etc.) • Intersectionality • BUT • Separation of gender and ethnicity • Where? Social movements defined it according to the ‘norm’ • Women’s movement – tendency to neglect ethnicity • Minority movements - tendency to neglect sex

5. Identity and Policy Implementation • Separation can further risk to create: • Polarisation between ethnic and gender identity • CHOICE: • Disempowered • Alienated from communities • => Inefficient policies

6. The Vulnerable Groups Survey, UNDP Europe and CIS • Integrated household survey • 9 countries of South East Europe • 2 groups IDPs/Refugees and Roma • Sample size app.1000 households per country (8 000 total) • Control sample of non-Roma living in close proximity with Roma and IDPs/Refugees • The case of Roma women

7. VGR- Case for Roma women income

8. Secondaryeducation

9. Unemployment, Serbia

10. Need of socio-economic data disaggregated by sex and ethnicity Problems related to ethnic statistics: Legal frameworks Fear Self-identification Problems related to gender statistics: Poverty measurements Employment Health and violence Challenges to data collection

11. The major challenge - “Who is Roma?” Compromise between self-identification and external identification with three levels of identification: Self-identification (reflected in the census) to identify the distribution and size of sampling clusters External identification (local activists, Roma experts, social workers) to identify the specific location of sampling clusters Potential respondents’ “implicit confirmation of the external identification” (identifying the individual respondents) Self-identification – Example of the Vulnerable Groups survey

12. Household as the unit of measurement (per capita) Assumes household as a unitary model driven by juste household head Economic income to the household does not by default mean an income for all household members Excludes intra-household distribution of resources Need to measure poverty on individual level Poverty measurements

13. Limitations: Multiple identities of minorities Trust towards interviewers Lack of gender and ethnic sensitive questions Suggestions: Multiple choice question on ethnicity Differentiate clearly between ethnic affiliation and citizenship Add questions on language, religion, parent’s ethnicity or country of birth Improve methods of fieldwork and involvement of minorities Data collection instruments - Census

14. Limitations: Fail to include representative sample of ethnic minorities With HBS/LFS drawn from the census that suffers from inaccurate population size data on minorities, under-representation of minorities in surveys is a direct consequence Suggestions: Boosters of respective minorities or separate minority samples Inclusion of sensitive questions Data collection instruments – HBS/LFS

15. Limitations: Ethnicity blind but gender sensitive Ethnic minorities opposed to introduce ethnic markers for fear of discrimination External identification problem in health, education and criminal registries Suggestions: Business registers could be engendered to provide information on entrepreneurship and agriculture Registries could be used for sampling surveys Data collection instruments - Registries

16. Limitations: Simple factors turn relevant: sex or ethnicity of the interviewer, the way a question will be asked, how the interviewer will be accepted by the respondent Interviews with women without presence of husband or other male household members Suggestions: Participation and involvement of the communities surveyed necessary at all stages of process Data collection instruments - Fieldwork

17. Collect gender and ethnically sensitive data together Statistical institutions need to have capacity to provide necessary guarantees on the privacy and use of the data Legal frameworks need to balance the protection of privacy and the need of anonymous data Existing data collection systems need to be sensitized to issues regarding ethnicity and gender by Sufficiently disaggregating existing data Developing adequate indicators that capture a wider context Complementing data collected with the household as a unit of measurement Involving the ethnic community in the collection of data Modifying existing questionnaires or creating new ones with emphasis on gender and ethnic sensitive questions Recommendations

18. Cooperation and partnership between data producers and users Standards for collected data (reliability, consistency, usefulness) Develop methodologies to complement registries data with unanimous survey-based instruments complementing ethnic dimensions to the specific topic studied Recommendations cont.

19. Thank you for your attention!