Sub-group meeting on Gender indicators by hcj


									                    4th Sub-group meeting on Gender indicators
                            Geneva, 27 September 2004
                              ECE, Statistics Division

                                     Minutes of the meeting

The meeting was held in ECE and attended by 27 experts from several international
organizations and national statistical offices. The list of participants is attached as Annex
The group reviewed and discussed proposals for indicators to monitor the achievement of
gender equality and empowerment of women, both under goal 3 and in the other MDGs.
At the previous meeting—held in Geneva on 27 May 2004—it was agreed that
responsible agencies would look at data available and present the results of their
assessment to the group, which would review the results and make the final decision on
what indicators would be proposed for inclusion in the official MDG list.
In particular, ILO was going to review data availability for calculation of regional
aggregates on several indicators related to employment in different sectors, including
formal/informal: WHO was going to suggest additional or alternative indicators to be
considered for national reporting and propose changes and/or additional indicators for
national reporting and in-depth analysis for the storylines and was proposing a new title
for the domestic violence indicator. Finally, IPU and UNSD were going to prepare
regional aggregates of data on ministries for 2000, and previous years if possible.1
The conclusions of the meeting as reported below were presented at the 6th Inter-agency
and Expert Meeting, which was held in Geneva on the following three days—29
September-1 October.
Employment indicators
Several proposals were reviewed for the indicator that should be used to monitor progress
in gender equality and women’s empowerment in the area of work. ILO representatives
introduced the proposed indicators and illustrated the related data limitations and
problems of interpretation.
First indicator, on employment by sector, composed of the following 3 dimensions:
    1. ―Women employment by sector (agriculture, industry, services) as percentage of
       women total employment‖, which provides the distribution of working women
       among the 3 sectors;
    2. ―Women employment by sector as percentage of total employment (all sectors
       and for both men and women)‖. Percentage points in each sector, if summed,
       give the share of women in total employment.

 The report of the Meeting of the Sub-group on Gender Indicators, held in Geneva, 27 May 2004, is
attached as Annex II.

   3. ―Share of women employment in each sector‖, which gives the percentage of
      women in each sector as percentage of total employment in the sector.
Second indicator, on employment by status:
   1. ―Women employment by status in employment as percentage of women total
   2. ―Women employment by status in employment as percentage of total
   3. ―Women employment as percentage of total employment by status in
Finally, ILO proposed an alternative indicator that would provide a comprehensive picture
of the situation of women in the labour force, by disaggregating data by different forms of
employment, including non-agricultural wage employment, informal employment
and agricultural employment:
   1. Share of women in total employment;
   2. Share of women in agricultural employment;
   3. Share of women in non-agricultural wage employment (current MDG indicator);
            a. Of which: 2.1 Informal wage employment;
   4. Share of women in non-agricultural self-employment;
            a. Of which: 3.1 informal self-employment;
It was stated that this indicator could be formulated in two different ways: (i) share of
women in each type of employment and (ii) women employment by type of employment
as percentage of women total employment.
With respect to data availability it was pointed out that, although the information on the
employment in the informal sector is available for more than 60 countries, substantial
effort would be required of the ILO to collect all information needed and that the
calculation of regional aggregates for 3(a) and 4(a) might not be relevant for all regions.
The meeting agreed that ILO would present the alternative proposal to the Inter-Agency
and Expert Group.
The representative from UNICEF presented the results of an analysis of gender
differentials in education based on data from household sample surveys (MICS and DHS)
and administrative records data provided by UNESCO Institute for Statistics. The
presentation and following discussion pointed out the following:
   1. The monitoring of gender differential/discrimination in education should be based
      on net rather than gross ratios. Net ratios give generally a more precise assessment
      of gender discrimination and they are considered to be a better measure of this
      phenomenon. In contrast, gross enrolment ratios seem to overestimate gender

   2. In order to have a more complete picture of the girls to boys differential in
      education, both school attendance and school enrolment should be used. School
      attendance is available from samples surveys while school enrolment data are
      derived from administrative data sources (enrolment registers).
The group agreed to present the following proposal to the Inter-Agency and Expert
1. Under goal 3, Net Attendance ratios should be used to monitor gender differentials in
   all three educational levels (primary, secondary and tertiary) with data obtained from
   household sample surveys; Net enrolment ratios should be used to monitor gender
   differentials in primary education only, given the limitations in data from
   administrative records for the other two levels.
2. Under goal 2, both Net and Gross Enrolment ratios should be analysed when
   assessing trends in enrolment in primary education.
3. UNICEF and UNESCO should continue to work together exploring and defining
   ways to optimize the use of different data sources for the monitoring of gender
   differences in access to education.
Reproductive health
A representative from the Millennium Project presented a proposal for the inclusion of
indicators to monitor the empowerment of women in the sphere of reproductive health
and their ability to control their reproductive life.
After discussion, the group agreed to propose the following modifications to the list of
MDG indictors, to be presented for approval at the 6th meeting of the Inter-agency and
Expert Meeting on MDG indicators:
      Move indicator 19c ―Contraceptive Prevalence Rate‖ from goal 6 (HIV/AIDS) to
       goal 5 (maternal mortality) as the group agreed this indicator is more relevant in
       monitoring the ability/capacity to avoid pregnancies than to protect individuals
       from HIV infection (of contraceptives methods, only condom use is a recognized
       reliable measure to prevent the infection).
      The inclusion of an additional indicator to monitor progress towards women’s
       empowerment in the sphere of reproductive health and their ability to control their
       reproductive life. Two alternatives were proposed and will be tested on the basis
       of empirical analysis on existing country data. They are as follows:
       *    The ―Proportion of Demand Satisfied (PDS)‖ expressed as the proportion of
            demand for family planning that is actually satisfied (ratio between the
            contraceptive prevalence rate--all methods--and the sum of contraceptive
            prevalence rate and unmet need). Data for this new indicator are available
            mainly from Demographic and Health Surveys for at least 75 countries (data
            compiled by the UN Population Division) and two time periods 1990-1995
            and 1996-2002.
       *    The ―Proportion of women currently 20-24 years old that were married
            before age 18‖. Data will soon be available from a longitudinal study that the
            UN population Division is currently carrying out.

Health indicators
WHO representatives emphasized the need to have data disaggregated by sex to the
larger extent possible. She reported on recent work being undertaken by the Department
of Gender, Women and Health (GWH) of WHO for the development of an advocacy
document on the importance of collecting and reporting data by sex geared at WHO,
other UN Agencies and Health Information Systems.
The importance of having data on access to drugs by sex (in particular antiretroviral
drugs, given the impact of AIDS in some regions) was especially emphasized.
More information on WHO position and suggestions in relation to the development of
gender indicators in health can be found in a document submitted by WHO after the
meeting and attached as Annex III.
Domestic Violence
As agreed at the previous meeting of the group in May, WHO proposed a new more
accurate definition for a possible indicator on domestic violence. The indicator is as
Measure of the burden of violence against women
Definition: Proportion of women who ever had a partner, 15-49 years old, who have ever
experienced physical violence by an intimate partner.
The differences with previous definitions are: (a) the focus on women aged 15-49 (to be
consistent with what is available from DHS data) rather than the larger age group, women
15-69; (b) the use in the denominator of only women who ever had a partner (thus at risk
of domestic violence) rather than all women. It was recognized that data availability at
this stage remains insufficient to produce the indicator for all regions and in two points in
time for trend analysis and that the focus should be on promoting adequate data collection
programmes. The sub-group pointed out the need for establishing a standard and regular
data collection process and agreed that the addition of a module on domestic violence in
all DHS and MICS surveys would be a good start.
Political participation
A representative reported that IPU and UNSD would work together to compile the data
series on women’s participation in government. The indicator would be ―share of women
in ministries‖ with possibly a disaggregation by type of ministry (women’s representation
is generally higher in social ministries than in economic and political ones). It was
pointed out that data are available from WISTAT for the years 1994 and 1998 for several
countries. The group agreed that IPU should explore ways to aggregate the data they have
for more recent years according to the WISTAT categories so that trend analysis could be

                                           Annex I
                                   List of participants:

                  Name                      Organization                          E-mail
1.    Ms. Francesca Coullare         DESA/UNSD        
2.    Ms. Francesca Perucci          DESA/UNSD        
                                     ILO/Policy Integration
3.    Ms. Heidi Solheim Johansen                      
4.    Ms. Angela Me                                   
5.    Ms. Karen Judd                 UNIFEM           
6.    Ms. Joann Vanek                UNIFEM           
7.    Mr. Jacques Charmes            IRD, France      
8.    Mr. Ralf Hussmanns             ILO              
9.    Mr. Shelly Abdool              WHO/GWH          
10.   Mrs. Lale Say                  WHO/RHR          
11.   Mr. Edilberto Loaiza           UNICEF           
12.   Ms. Lucia Fort                 World Bank       
13.   Ms. M. Eugenia Gomez Luna      INEGI/Mexico     
                                     Millennium Project –
14.   Mr. Stan Bernstein                              
                                     Task Force 3
15.   Ms. Yianna Lambrou             FAO              
16.   Mr. San Sy Than                NIS Cambodia     
17.   Mr. Atiqur Rahman              IFAD, Rome       
18.   Mr. John Kahimbara             Statistics South Africa
19.   Ms Reiko Tsushima              ILO              
20.   Ms. Linda Wirth                ILO/Gender Bureau
21.   Ms. Sophia Lawrence            ILO/Bureau of Statistics
22.   Mrs. Valentina Stoevska        ILO/Bureau of Statistics
23.   Ms. Francesca Comincini        UNECE            
24.   Mr. Enrico Bisogno             UNECE/Stat Div.  
25.   Ms. Sylvie I. Cohen            DESA/DAW         
26.   Ms. Lynn MacDonald             UNDP/PG          
27    Mr. Ahmed Hussein              UNESCWA          

                                       Annex II
 Meeting of the UN Inter-Agency and Expert Meeting on MDG Indicators, Gender
                  Indicators Sub-group (Europe-based agencies)
                                  ECE, Geneva
                                  27 May 2004

The meeting was held to discuss indicators related to monitor the achievement of gender
equality and empowerment of women, both under goal 3 and in some of the other MDGs.
The meeting began by approving the draft agenda. Participants requested the inclusion of
the following:
      under the item ―Proposed indicators for goal 3‖, the inclusion of a discussion on
       work in the agricultural sector, and an item on ―other indicators for review‖;
      under ―Proposed priority indicators for inclusion in national progress reports,
       etc.‖, the inclusion of indicators on access to technology.
The revised agenda as approved by the group is attached.
In response to questions from the group, Francesca Perucci explained the links between
this group and other groups working on engendering the MDGs (the task force led by the
WB, the MP, etc.). Also, the process for the indicators to be considered for inclusion in
2005 was explained and it was pointed out that an opportunity to propose new indicators
will occur after the Millennium Summit+5 meeting in 2005. If the Gender Indicators
Working Group reaches a consensus about new or alternative indicators, they will make a
proposal to the full Inter-Agency and Expert Group on MDG Indicators, and if the full
Meeting approves this proposal, they will send it forward to the Secretary General’s for
consideration in 2005.
Before moving to the discussion on the specific indicators, the group also talked about
the importance of using for international comparison, all existing national sources, which
are not sufficiently and timely incorporated in international data series.
There was also a discussion on the importance of incorporating in the analysis presented
in the MDG report, other issues that are not suitable to be addressed by official
indicators–such as for instance the question of quality of education.
Proposed indicators for Goal 3

Indicators on employment
The current indicator is ―share of women in wage employment in non agricultural
sector‖. The supplementary indicator proposed in various fora, including the meeting of
the sub-group on gender indicators in New York, and the Task Force 2 of the Millennium
Project, is an indicator on informal employment.
ILO representatives however, pointed out that the indicator on the informal sector was
difficult to assess and therefore might not be necessarily the best choice. They also
pointed out that data availability remains a major problem and wanted the reactions of the

group on further possible alternatives. They presented statistics based on a review of
current data availability – geographical and time coverage – for two other indicators:
       1. Women’s share of total employment by occupation, looking at administrative
          and managerial category (data for 96 countries, only 2 in Sub-Saharan Africa).
       2. Women’s share of total employment by status in employment, looking
          specifically at own-account workers, although other statuses should also be
          shown (data for 60 countries, none in Sub-Saharan Africa),
The ILO representatives pointed out that, in spite of the problems with the different
versions of the ISCO classification used, the first proposal is (i) very indicative, it shows
the male to female ratio in the group of administrative and managerial occupations over
time, and (ii) it is easily interpreted. The second proposed indicator, on the other hand, is
more difficult to interpret, because the classification of self-employment (own-account
workers) differs across countries, especially as concerns the distinction between own-
account and unpaid family workers-with generally more impact on the female labour
force. As a result, the aggregation of data at regional and global level may not be
They also proposed looking at wage sex-based differentials as another alternative.
 The discussion that followed highlighted the shortcomings of using these further
proposed indicators for two sets of reasons: the limited geographical coverage, excluding
the poorest countries and Sub-Saharan Africa, which in fact should be the focus of the
attention, on the one hand; and the exclusion of a large part of women workers who are
either in other sectors –such as agriculture–or hold informal jobs and/or perform unpaid
work. It was also pointed out that women’s share in self-employment, given that the
category is broader, is probably available for a larger number of countries and might also
be used as a proxy for informal sector (see below).
ECE pointed out that another useful indicator for gender-based analysis in the area of
work and employment is work participation by presence of young children in the
household. This indicator was one of the four already proposed by ILO, given that the
necessary information exists for all countries that regularly conduct a Labour Force
Survey, even though currently no such data are made available in international sources.
After extensive discussion on problems of interpretation, coverage, possibility of
aggregating the data at the regional level and of looking at trends, there was agreement
that the data at their current status do not allow the calculation of the indicators proposed
for global and regional monitoring. Also ILO clarified that there are no methodologies
that would allow extending the coverage through estimates, given the complete absence
of basic information for a large number of countries, especially in Africa. It was therefore
decided that the indicators should not be included for consideration in the 2005 proposal.
However, given their importance and relevance to gender they should be kept in the other
tiers of the proposed list.
The group then discussed the proposal on an indicator on the informal sector that
originated from the work of the Millennium Project task force on gender and was
reviewed by the meeting in NY on May 17. ILO distributed a table indicating the
number of countries with official statistics on employment in the informal sector, by

region and geographic coverage. The main question discussed was whether to limit the
analysis to the relatively small number of countries that have data on the informal sector
for two points in time (34 countries to date), or consider self-employment as a proxy for
informal sector and obtain data for a larger number of countries.
The representative from FAO raised the question that the current indicator on wage
employment completely excludes agriculture, which in many developing countries is a
significant part of women’s employment. Also, the alternatives proposed, both by ILO
and by the MP task force and the sub-group that met in NY–an indicator on the informal
sector–all look at non-agricultural work, neglecting a large part of the population, in
particular women.
The discussion that followed highlighted that, given the large presence of women in
agricultural work and subsistence work, indicators in the area of work should reflect the
situation of women in this sector as well.
Problems of interpretability, however, exist for all the proposed indicators. There are no
clear indications of what should be considered a positive trend and what should be
considered desirable for women’s situation in the labour force. In the case of wage
employment in non-agriculture, for instance, an increase in the share of women may not
necessarily coincide with an improved situation as concerns pay, working conditions, etc.
An increase in another status in employment might be more desirable (e.g. becoming
self-employed as employer). Similarly, when looking at women in agriculture, an
increase in the share of women in wage employment is not necessarily an improvement in
their working opportunities, and rather a shift from agricultural jobs to non-agricultural
and better-paid jobs might be more desirable. In the case of the informal sector, the
interpretation is even more complex as shifts from formal to informal and vice-versa can
mean something very different in different countries and regions, depending on
opportunities available, legislations, the levels of income generated by the different
activities, etc.
Nevertheless, the group agreed that these indicators are still the ones that might give a
comprehensive picture of the situation of women in the labour force and that agriculture
and the informal sector cannot be excluded from the analysis.
The following indicators were therefore listed to be considered for possible inclusion
under goal 3:
     1.    Labour force participation rates by main industrial sectors (according to ISIC,
           agriculture, industry and services) (data already available)
     2.    Share of women in wage employment in non-agricultural sector and in wage
           employment in the agricultural sector
     3.    Share of women in the informal sector
     4.    Share of women in self-employment
Before a final decision can be made, ILO will need to provide a review of available data
and show what the analysis of the data at the regional level and over time can reveal.

Health indicators
The representative from WHO pointed out that there is currently no sub-group of the
Inter-agency and Expert Group for health indicators. Therefore, the responsibility for
making recommendations on how to incorporate a gender dimension in all MDG health-
related indicators lies on this group.
Specifically, the agreement was that the group should formally recommend WHO and
other responsible agencies, to provide data by sex where they exist or launch initiatives to
collect and report data by sex where it is not done yet in the following areas: infant and
child mortality, child malnutrition, TB, HIV/AIDS. WHO is also to provide the analysis
of the gender dimension in the various targets related to health, as done in the past, for
inclusion in the storylines to be used in the preparation of the 2005 comprehensive report.
Finally, the gender specialist at WHO will review the current list of proposed indicators,
“Proposed Gender-Sensitive Development Indicators to Monitor Progress toward MDG
3‖, in the other tiers (B, C, and ―Indicators for national reporting‖) and propose changes
and/or additional indicators for national reporting and in-depth analysis for the storylines.
They will also review alternatives for an indicator to be included in one of the above-
mentioned tiers, on adult mortality by sex.
The discussion also focused on the question of access to drugs, both in general as it is
under Goal 8, and in relation to access to anti-retroviral drugs. It was felt that access to
drugs should be monitored more closely and that given the enormous number of people
already dying of AIDS in the African countries, it is not sufficient to focus on knowledge
and prevention, but that urgent action should be on the side of saving the lives of those
already infected. Since access to drugs – as is the case of access to health services- has a
strong gender dimension, the group recommends WHO to do further work in this area
and propose ways of collecting the data (such as for instance through questions to be
included in DHS, MICS, etc.)
Violence indicator
The representatives WHO provided an overview of data availability, problems of
interpretation and possible ways forward. Currently, data are available from surveys such
as DHS and Crime Victimization Surveys, for about 40-50 countries. Problems of
interpretation relate to the fact that a higher percentage of women victims of violence
may indicate an actual increase as well as the fact that more women than before report the
crime. Another important aspect raised during the discussion is that the indicator is on
women only and therefore does not allow a gender-based analysis. The group
recommended that the indicator be rephrased specifying that data are presented by sex.
The final recommendations of the group were that the overall idea of including an
indicator on domestic violence is a valid one. However, it is premature to include the
indicator among those proposed for review in 2005, given that data are still very limited
and problems of comparability and interpretability do not allow any trend and/or regional

Indicators on Access to technology (Indicators 47 and 48, Target 18 under Goal 8)
The representative of ITU reported to the group that data on Internet users are now by sex
are now available for about 60 countries. The data are from surveys carried out by
national agencies as well as market research companies.

Indicators on Political participation
The representative of IPU reviewed the various proposals for additional indicators on
women’s participation in decision-making.
Both IPU and International IDEA have made an attempt to compile data on voting
participation, by requesting countries to provide voters turn out by sex. Unfortunately,
very few countries (electoral commissions) disaggregate voter turnout by sex.
On the share of women ministers, IPU will be able to compile data series for the years
2000 and 2005. Data was collected in 2000 on senior and junior ministers, and on
parliamentary secretaries for a poster on women in politics produced with UNDAW.
This exercise will be repeated in 2005. Some of the older series were in the UNSD
database Wistat. UNSD will explore the possibility of combining the series to allow a
trend analysis over a longer period of time. Data will cover ministers and junior ministers
by portfolio. The group agreed that the two levels of ministers will be used as a
supplementary indicator for the storylines, without proposing it as a new official MDG
There was also a discussion on whether to consider an indicator of political participation
at sub-national level. Given that this indicator is relevant and in some settings women’s
political participation at this level is important and has a great impact, it was decided to
consider it in tier C. The indicator, however, is not suitable for regional and global
The group also discussed the importance of analysing the involvement of women in
policy making specifically related to gender issues. One way to look at this is to show
whether or not there are committees on gender policies.
Conclusions and follow-up
The responsible agencies will look at data available and prepare the analysis to be
presented at the next meeting of the sub-group to be held in Geneva, at ECE, in
September this year. The group will review the results and analysis and make the final
decision on what indicators will be proposed for inclusion in the official MDG list. The
decision will then be reported to the plenary discussion at the MDG Inter-agency and
Expert meeting, to be held immediately after.
In particular, the following work will be completed for circulation in time for other
members to review the material before the next sub-group meeting:
ILO            Review of data availability and calculation of regional aggregates for:
                      Labour Force Participation rates by sector (agriculture, industry
                       and services)

                Informal sector
                Self-employment
                Wage employment in agriculture
                Wage employment in non-agriculture

WHO       Suggest additional or alternative indicators to be included in tiers B, C,
          and ―Indicators for national reporting‖and propose changes and/or additional
          indicators for national reporting and in-depth analysis for the storylines;
          Prepare a note with recommendations on data collection and reporting by
          sex on health indicators covered under the other goals (Goal 1, Child
          malnutrition; Goal 4, Child mortality, Infant mortality, Measles
          immunization; Goal 6, HIV/AIDS, TB and Malaria)
          Propose a new title for the domestic violence indicator
IPU and
UNSD      Prepare regional aggregates of data on ministries for 2000, and previous
          years if possible.

          Make suggestions for an indicator on political participation at sub-national

          Make suggestions for an indicator on involvement of women in policy
          making specifically related to gender issues (such as for instance, whether
          or not there are committees on gender policies)

                                  Annex III
              Proposal for additional and alternative indicators
                      on health and gender in MDG 3
 by WHO, Department of Gender, Women and Health (GWH), Integrating Gender
                         into Public Health (GPH)

I. Recommendations on data collection and reporting by sex on health indicators
covered under the other goals
It is a widely recognised fact that the single most jarring barrier to undertaking gender
analyses of health data, and uncovering gender inequalities in health, is:
a) the absence of sex disaggregated data;
b) data collected by sex at community/local level but then aggregated at
national/international level, and
c) data collected by sex but not analysed.
GWH/WHO is currently developing an advocacy document on the importance of
collecting and reporting data by sex geared at WHO, other UN Agencies as well as
Health Information Systems.
       It is foreseen that GWH/WHO will link up with other UN Agencies to lead a
campaign that addresses the ―counting‖ of women’s concerns in health systems and
information. Without accounting for the differential risk factors and health outcomes that
women and men experience, proper analyses of the interaction between gender and other
determinants of health remain weak. The Gender Indicators sub-group is asked to
strongly recommend and call for sex disaggregated data across the MDG indicators.
Special attention should be drawn to measuring issues of access to essential medicines
and services (including anti-retroviral therapy – ART).

II. Additional or alternative indicators to be included
(Please note that additional or alternative indicators for concerning reproductive health
are included in the proposal put forth by Dr. Stan Bernstein, supported by the RHR
In reviewing the additional or alternative indicators to be included in tiers B, C and in
Indicators for national reporting (of the prioritized list from May 2004), GWH/WHO has
sorted through the many possible health indicators that are not already covered under
other MDG indicators (regardless of associated specific goal or target) to determine those
most able to reflect important markers for progress towards gender equality and
women’s empowerment.

 Suggestion 1
 Measure of the burden of violence against women
 Definition: Proportion of women who ever had a partner, 15-49 years old, who have
 ever experienced physical violence by an intimate partner. [Sources: Special surveys]

It was suggested at the last meeting that WHO propose more appropriate phrasing for an indicator
that could measure VAW. The above indicator is based on the definition of intimate partner
violence as defined through the WHO Multi-Country Study on Women’s Health and Domestic
Violence and other similar surveys. The age range of 15-49 years old is used as it covers the
reproductive years of the life cycle which allows for pregnancy-related violence to be included.
This age range also reflects the fact that violence tends to occur early in an intimate relationship
and is the range used in most VAW surveys to date (i.e., WHO, CDC, DHS). The definition of
―intimate partner‖ is variable as it depends on the country or context. The WHO Multi Country
Study currently adapts the definition to the given context. For example, some countries define
intimate partner as spouse or husband while others define it as cohabitating and/or non-
cohabitating partner.
In terms of expected data development beyond 2005, there is consensus amongst researchers
working on VAW to identify and/or develop an indicator that will capture trends in VAW as a
measure of gender equality and to account for the health consequences of VAW. More recently,
a debate has begun regarding which indicator should be used. Prevalence of VAW is one such
indicator. The WHO Multi Country Study on Women’s Health and Domestic Violence uses a
Violence Against Women instrument to measure emotional, physical and sexual violence against
women by an intimate partner. With respect to physical violence, the instrument has a 6-item
subset of questions (and can be sent electronically upon request). GWH can provide prevalence
data for the countries in which this is available.

 Suggestion 2
 Measure of Early Marriage
 Definition: Proportion of 20-24 year olds married before the age of 18, by sex.
 [Sources: DHS, national surveys]

Early marriage tends to lead to early childbearing, which can be dangerous for the immediate and
long-term health of women. Also, it is associated with large intra-spousal age differences and low
status/power for the wife in the new family, which can in turn negatively affect health. This intra-
spousal age difference (coupled with the lack of choice of marriage partner where arranged
marriage is the norm) and its implied power relations may also be a factor of forced sex within
marriage, or other forms of sexual coercion in marriage (defined as deception, verbal threats or
psychological intimidation to obtain sex, attempted rape and forced penetrative sex. Forced
marital sex can also be accompanied by physical or emotional violence.
This indicator is not only a good measure of progress towards gender equality and women’s
empowerment in terms of intra-personal relationships and detrimental health outcomes. Early
marriage often significantly reduces young girl’s and women’s access to education and/or gainful

employment opportunities under safe working conditions. In this sense, this indicator is
privileged for its ability to measure progress in achieving MDG 3.

  Suggestion 3
  Measure of Son Preference
  Definition: Ratio of female to male live births. [Sources: Census data, DHS]

Included among the health indicators in the May 2004 prioritized list is life expectancy (LE).
However, LE - even if disaggregated by sex - does not adequately measure aspects of gender
equality and women’s empowerment as it masks the fact that while women may live longer than
men, they tend to be sick more often (and have more chronic conditions than men) – especially in
later years. HALE (health adjusted life expectancy) and LHE (expected lost healthy years at
birth) are not feasible indicators either, due to difficulties in data calculations.
The proposed indicator of ratio of female to male live births could potentially account for
occurrences of female infanticide and sex selection – practices that are inextricably linked to
gender inequality and the value placed on boy versus girl children.

For more information, please contact:
     Dr. Claudia Garcia-Moreno, Coordinator, Department of Gender, Women and Health)
     Ms. Shelly Abdool, Technical Officer, Department of Gender, Women and Health
     Dr. Henrica Jansen, WHO Multi Country Study on Women’s Health and Domestic Violence


To top