Upload
eugene-milton-wilkerson
View
221
Download
0
Tags:
Embed Size (px)
Citation preview
Methods to measure discrimination
Initial ideas to develop a toolbox
Quantitative Assessment of Human Rights and Development Workshop
Oslo, May 2010
Eitan FelnerIndependent Consultant
Introduction
• Part of a larger project – toolbox to measure ESCR• General framework
• Sur article: ‘A new frontier in economic and social rights advocacy? Turning quantitative data into a tool for human rights accountability’ (CESR)
• Toolbox for UNDP Oslo governance center
• Progressive realization according to available resources
Article: ‘Closing the ‘Escape Hatch’: A Toolkit to Monitor the Progressive Realization of Economic, Social, and Cultural Rights’ -
• Work in progress (initial suggestions)
Basic premise
- Discrimination is a complex issue• Direct and indirect discrimination
-Across various stratifiers (prohibited grounds) – gender, ethnicity, etc
- in household, communities, states
- Multiple forms of discrimination requires multiple tools to measure discrimination
3-step Methodological framework
#1 Identying disparities in enjoyment of ESCR
#2 Identifying causes for disparities in enjoyment of ESCR
#3 Measuring discriminatory policies
Effect
Cause
Methodological steps
Step #1 – Interpreting unequal enjoyment of ESCR
• Basic premise: Extent to which disparities amounts to discrimination depends on specific indicator and stratifierRules of thumb: gender disparities in enjoyment of ESCR that benefit boys/men typically related typically related to discrimination against girls/women at the household, community and/or or state level.“where no plausible biological reason exists for different health outcomes, social discrimination should be considered a prime suspect for different and inequitable health outcomes.” WHO 2007a(e.g. Save the Children – gender disparities in child mortality in China and India)
Caveat: not all health differences between sexes reflect gender discrimination.e.g. instance, differences in birth weight between girls and boys: boys universally tend to weigh more than girls at birth.
Measurement adjustments: women on average live longer lives than men. Amartya Sen suggestion: easuring gender equality in health outcomes in terms of ‘shortfalls’ from the optimal value that each sex can respectively attain
Angola Sub-Sahara Africa
Lower-Middle Income
(12) (22) (270)
Lifetime risk of maternal deaths, 1 in:
= Maternal deaths
Step #1 – Interpreting unequal enjoyment of ESCR Example #1
Contents
Selection of countries
Data and figures
Others
NIGER (44.5%)
Total boys (59.6%)Rural boys (53.4%)
Urban boys (85.4%)
Urban girls (62.9%)
Total girls (30.0%)
Rural girls (22.5%)
80%
60%
40%
20%
DJIBOUTI (35%)
MALAWI (54%)
GUINEA (64%)
EL SALVADOR (84%)
Niger - Wide gender and regional disparities in education
Primary Completion Rate by area of residence, 2006
Step #1 – Interpreting unequal enjoyment of ESCR Example #2 – compounded inequalities
#1 Interpreting ethnic disparities in the enjoyment of ESCR
Rule of thumb: ethnic disparities in enjoyment of ESCR: associated with underlying discriminatory policies and/or are the accumulative result of historic discrimination
Caveat: Ethnic disparities in enjoyment of some rights such as health may sometimes be related to factors beyond state’s actions or inactions cannot be attributed to past or present patterns of discrimination.
• health problems due to genetic reasons may be common among certain ethnic groups and not others
• natural environment where an ethnic group is concentrated may be more prone to some time of diseases (e.g. malaria) than where other groups live e.g.: reportedly, one of several reasons why the Kikuyu in Kenya had a much lower under-5 mortality rate than non-Kikuyu (Brokerhoff and Hewett 2000)
•cultural habits of some ethnic groups that may contribute to poorer enjoyment of ESCR. (e.g. food habits of some ethnic group may contribute to malnutrition among children).
#1 - Interpreting ethnic disparities in the enjoyment of ESCR (cont)
Nevertheless: need to focus on gov’t response to disparities rather than on who/what’s responsible for creating disparity
Key question: has gov’t failed through its inaction to respond adequately to the conditions that created, exacerbated or perpetuated such disparities? (including affirmative action measure to overcome historical discrimination)
Step #2 - Identifying causes for disparities in enjoyment of ESCR
High incidence of girls out of primary school
Parents refusal to send girls to
school
Parents can’t afford school fees
Cultural belief and practices
Teachers are often absent
Government’s responsibility
School too far away
Gov’t forbids girls to attend
school
Poor quality of teaching
Supply Demand Factors outside sector
Step #3 – Measuring discriminatory policies – Direct discrimination by state
.-
Type #1. Discriminatory laws and social institutionsExamples:
discriminatory laws and social institutions (minimal age for marriage, property rights, differential minimum wages for women and men, requiring the permission of father or husband before a woman (even ifshe is a legal adult) can apply for a loan, for a development programme, for education, reproductive health services, etc)
Type #2 Lack of adequate resources for specific needs (example in next slide)
Type #2 - Lack of adequate resources for specific needs
Pakistan: Low commitment to public spending on primary education despite low levels of girls’ school enrolment
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
16.00
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
90.00
Pakistan South Asia
Female net enrolment rate, primary
Public education expenditure, % of Gov-t spending
Female Net Enrolment Rate and Public Expenditure on Education
Step #3 – Measuring discriminatory policies – Direct discrimination by state
Sources: ENCOVI 2006, Rubio and Salanic (2005) and UNDP Guatemala 2005
MethodComparing multiple data sets
Poverty IncidenceTeachers’ Reading
Test ScoresConcentration of
Indigenous People
Dept. Poverty Dept. Score Dept. % Pop. IndigenousQuiché 81 Sacatepéquez 72.6 Totonicapán 98%Alta Verapaz 78.8 Guatemala 66.5 Sololá 96%Sololá 74.6 Chimaltenango 66 Alta Verapaz 93%Totonicapán 71.9 El Progreso 61.4 Quiché 89%Huehuetenango 71.3 Retalhuleu 60.5 Chimaltenango 79%Baja Verapaz 70.4 Petén 60.5 Huehuetenango 65%San Marcos 65.5 San Marcos 60.2 Baja Verapaz 59%Jalapa 61.2 Zacapa 59.9 Quetzaltenango 54%Chimaltenango 60.5 Jalapa 59.8 Suchitepéquez 52%Chiquimula 59.5 Chiquimula 59.3 Sacatepéquez 42%Santa Rosa 57.9 Escuintla 58.8 San Marcos 31%Petén 57 Suchitepéquez 57.4 Petén 31%Suchitepéquez 54.7 Quetzaltenango 56.8 Retalhuleu 23%Zacapa 53.9 Baja Verapaz 56.2 Jalapa 19%Retalhuleu 50.4 Jutiapa 55.6 Chiquimula 17%Jutiapa 47.3 Totonicapán 54.2 Guatemala 14%Quetzaltenango 44 Huehuetenango 53.5 Escuintla 7%El Progreso 41.8 Santa Rosa 52.5 Jutiapa 3%Escuintla 41.4 Sololá 51.4 Santa Rosa 3%Sacatepéquez 36.5 Quiché 51.2 El Progreso 1%Guatemala 16.3 Alta Verapaz 50.9 Zacapa 1%
Type #3 – Measuring inequality in quality of service provision - Assessing whether marginalized children are being taught by the least qualified teachers
Step #3 – Measuring discriminatory policies – Direct discrimination by state
Laws, policies or practices which appear neutral at face value, but have a disproportionate impact on the exercise of rights by some social group, such as women, indigenous peoples or an ethnic minority).
Gender blindness: non-recognition of the ways in which control over productive assets, division of labour, decision-making, physical mobility among other factors, are biased against women.
Examples:- Not building separate bathrooms in schools for girls and boys
- Seemingly gender neutral policy of not having schools in every village may - disproportionately affects girls, because of highly influential cultural norms in some countries related to the restriction on women’s mobility (e.g. Pakistan)
- Requiring a birth registration certificate for school enrolment may discriminate against ethnic minorities or non-nationals who do not possess, or have been denied, such certificates.
Step #3 – Measuring discriminatory policies – Indirect discrimination by state
Example: North India - preferential treatment of boys and neglect of female children in the intra-household allocations related to health care, nutrition and related needs, are contributing factors to perduring gender disparities in child mortality in India in favor of boys (Dreze and Sen 2002)
Step #3 – Measuring discriminatory policies – Discrimination within household