The fundamental principle of the 2030 development agenda is to leave no one behind. Achieving real inclusion – and monitoring progress – will require a significant improvement in the availability of data disaggregated by gender, age, and other important attributes. The indicators proposed for the Sustainable Development Goals (SDGs) show a strong commitment to expanding the global monitoring effort to those groups previously left out or poorly represented, including girls and women. The size and scope of the indicator set also reveals many measurement challenges that countries will face as they develop statistical systems to accommodate the demand for new data. Meeting this demand will require political commitments from national, regional, and international stakeholders, increased resources, and new tools and processes. And while the need for new data is urgent, we must be realistic and recognize that building the necessary infrastructure will take time as well as money.
Not only do the Ready to Measure indicators provide important information about the economic status and welfare of women, they can kick-start the measurement of a gender baseline for the SDGs.
Our Ready to Measure study (published September 2015; reproduced March 2016) demonstrated that while many of the proposed gender-related indicators are challenging to collect, there are 20 indicators (16 SDG indicators and 4 additional complementary indicators) that are ‘ready to measure’ with internationally agreed definitions. They are produced with available data collection instruments and already have good coverage. With the adoption of the final list of proposed SDG indicators (March 2016), we have compared our original Ready to Measure indicator list with the SDG indicators, included as Annex 3. The good news is that we identified significant overlap between these lists –16 Ready to Measure indicators are the same or similar to those in the proposed final list of SDG indicators. Not only do the Ready to Measure indicators provide important information about the economic status and welfare of women, they can kick-start the measurement of a gender baseline for the SDGs.
Going forward, our goal is to encourage and assist the start of SDG baseline data collection immediately using these 16 indicators. We plan to continue this work and keep an inventory of the Ready to Measure indicators along with other SDG gender indicators. We will estimate costs of traditional and alternative methods of data collection and highlight emerging opportunities and examples of best practices.
We are open to and welcome joining forces with other partners, experts, and national agencies to pilot Ready to Measure indicators and motivate a strong start to gender data collection in the era of SDG implementation.
Background
The transformative agenda and ambitious scope of the Sustainable Development Goals (SDGs) has resulted in a long list of indicators that will need to be monitored at national, regional, and global levels. Many of these indicators are ‘aspirational’ and will take time and significant resources to produce. But the clock is ticking, and intensified collection of SDG baseline data needs to get solidly off the ground in 2016. This note proposes twenty indicators to launch baseline data collection on the situation of women and girls. In many countries, consistent, historical series for these indicators already exist, their sources are well established, and data collection can start or resume immediately. Countries where significant gaps remain should receive priority attention.
This list of proposed indicators is intended as a contribution to ongoing work of the gender community, the UN system, and member countries. It is not intended to supplant the recommendations of UN Women, the IAEG-GS, or the Open Working Group. It is complementary to the efforts of the Global Partnership for Sustainable Development Data to identify ‘early wins’ that can be implemented in the near term. By initially focusing on a limited but available set of indicators while work continues to develop new indicators and expand disaggregation of other indicators, continuity with historical series and a strong base for SDG monitoring can be established.The indicators were selected using UN Women’s list of 47 indicators to monitor gender equality and women’s empowerment in the SDG framework and the master list of 52 indicators identified by the Interagency and Expert Group on Gender Statistics (IAEG-GS). Many of these indicators also appear on the list compiled by the Bureau of the UN Statistical Commission (UNSC). This list was purposely narrowed down to a manageable number of universally applicable indicators following the criteria discussed below. Whenever feasible, these indicators should be further disaggregated by income level, race and ethnicity, and disability to track inclusive progress for women and girls. Subnational disaggregation is also encouraged. Other indicators recommended for monitoring the SDGs should also be disaggregated by sex, where possible. In addition, indicators should be identified to capture issues that are not global but are specific to certain regions or countries, or sub-populations of women and girls (such as female genital mutilation).
Proposed list of Ready to Measure gender indicators
Goal 1: End poverty in all its forms everywhere.
- Ratio of women to men (aged 15 and above) in households living under the international or national poverty line6.
Goal 2: End hunger, achieve food security and improved nutrition, and promote sustainable agriculture.
- Prevalence of stunting (low height-for-age) in female children under 5 years of age, and ratio of female to male under-five
- Prevalence of anemia in women of reproductive age (15 to 49), pregnant and non-pregnant.
Goal 3: Ensure healthy lives and promote well-being for all at all ages.
- Maternal deaths per 100,000 live
- Female under-five mortality rate and ratio of female to male under-five mortality7.
- Number of female new HIV infections per 1,000 susceptible population (by age and key populations), and ratio of female to male new HIV
- Adolescent birth rate per 1,000 women in that age group, by age of the mother (15-19).
- Percentage of reproductive age women (15-49) using modern
Goal 4: Ensure inclusive and equitable quality education, and promote lifelong learning opportunities for all.
- Percentage of girls/young women aged 3-5 years above official age for last grade of each level of education (primary, lower secondary, upper secondary, and tertiary) who have completed the level, and female to male
- Share of women aged 15-49 who use the computer and/or internet at least once a week, and every day
Goal 5: Achieve gender equality and empower all women and girls.
- Percentage of women aged 20-24 who were married or in a union before age 18 (child marriage).
- Proportion of seats held by women in national
- Share of women among mobile telephone
Goal 8: Promote sustained, inclusive and sustainable economic growth, full and productive employment and decent work for all.
- Proportion of young women who are idle (women 15-24 who are not employed and not in school and not looking for work).
- Females employed as a ratio of the working-age female population (15 to 59), and female to male
- Proportion of employed who are own-account (self-employed) workers by sex of
- Women’s share of non-agricultural wage
- Percentage of adult women with a formal financial account or personally using a mobile money service in the past 12 months, and female to male
Goal 10: Reduce inequality within and among countries.
- Growth rate in adult women’s share of household earned income among the bottom 40 percent of the population, relative to that for all adult
Goal 16: Promote peaceful and inclusive societies for sustainable development, provide access to justice for all and build effective, accountable and inclusive institutions at all levels.
- Proportion of female children under 5 whose births have been registered with a civil authority, and female to male ratio.
How much will it cost?
All but two of the indicators listed above can be constructed from population censuses and household surveys regularly administered in many countries8. The principal survey sources in low- and lower-middle income countries with individual-level data are the Demographic and Health Survey (DHS), supported by USAID; the Multiple Indicator Cluster Survey (MICS), supported by UNICEF; Living Standards and Measurement Study (LSMS), supported by the World Bank; and the Labor Force Survey (LFS), supported by the International Labor Organization (ILO). The Global Financial Inclusion Database (Findex), which is housed at the World Bank and was conducted as part of the Gallup World Poll in 2011 and 2014, covers individual nancial outcomes across low-, middle-, and high-income countries9. Finally, country-speci c household welfare surveys, household budget and consumption surveys, as well as population and agricultural censuses are also an important resource, albeit not necessarily comparable across countries. Support does not mean that the surveys are costless to the countries, but there is often signi cant technical assistance provided, and, in the poorest countries, grants are available.
The Sustainable Development Solutions Network (SDSN) has published an assessment of the cost of producing 100 of the most important SDG indicators, including many of those listed here. For the 77 low- and middle-income countries that qualify for concessional lending from the World Bank’s International Development Association (IDA), the cost of conducting the household surveys is estimated to be $134 to $173 million a year. Censuses over the 15-year SDG period will cost $320 million, and improvements to civil registration and vital statistics (CRVS) systems are estimated to require $220 million. Although these are substantial amounts, the additional cost of producing the indicators discussed here is likely to be quite small. In some cases survey questionnaires may need to be revised to include items needed to construct the indicators, but recent surveys by DHS and MICS already include the necessary questions. For a few indicators, such as indicator 1: the ratio of women to men (aged 15 and above) in households living under the international poverty line; and indicator 19: growth rate in adult women’s share of household earned income among the bottom 40 percent of the population, relative to that for all adult women, the methodologies for calculating the indicators from existing and future surveys will need to be developed and implemented. One additional concern is for the continuation of the World Bank’s Findex surveys, which produce information on participation in the formal financial system.
Because most of the required surveys are already conducted under the auspices and with financial support from bilateral or multilateral agencies, the marginal cost of producing these indicators (and many more) is only the cost of conducting censuses and surveys or increasing their frequency in countries with data gaps. Recent estimates by Chandy and Zhang suggest these costs, assuming continuation of current survey programs, will be quite small.
Annex 1: Notes on sources and uses
This annex provides brief metadata for the proposed indicators, including notes on their definition and data sources. Other indicator lists that incorporate these or similar indicators are identified as: MDG – Millennium Development Goals Indicators (website). See UNSC-B – Technical Report of the Bureau of the UN Statistical Commission. See IAEG-GS – Interagency and Expert Group on Gender Statistics report to the 2013 UN Statistical Commission. See UN-Women – Minimum set of gender indicators. See Empowering Women – Monitoring Gender Equality and Empowerment of Women and Girls – 2030 Agenda.
1. Ratio of women to men (aged 15 and above) in households living under the international poverty line.
This indicator is readily measurable from available data collected through income-expenditure surveys. Other measures of female poverty, such as the poverty status of female-headed households, would require additional data collection and an agreed methodology for establishing the head of household. However, because of systematic variations in the composition of households, this indicator is not suitable for cross-country comparisons. For example, in countries where there is a strong preference for male children to the extent that there are “missing females” and very skewed sex ratios, the indicator will show lower numbers of women below the poverty line than in countries and regions where women are in reality better off. Therefore it should only be used to track changes over time within a country. Because international comparisons are not involved, the indicator could also be measured at the national poverty line. A similar indicator could be calculated using measures of relative wealth from DHS surveys.
Data on household consumption and composition are available in all LSMS and other country- specific household welfare surveys. Computation of this indicator at the international poverty line requires a PPP (purchasing power parity) conversion factor for consumption and a domestic consumer price index. Computation at the national poverty line may require a consumer price index to adjust for price level changes between the survey year and the base year of the poverty line.
This indicator is included in the UN Women report as the “Ratio of women to men (aged 25-59) in households living under $1.25 per capita (PPP),” and, separately, “Ratio of women to men (aged 25-59) in households living under the national poverty line.”
2. Prevalence of stunting (low height-for-age) in female children under 5 years of age, and ratio of female to male under-five stunting.
Coverage is already quite high among the relevant survey categories (health/DHS and MICS). Some older DHS surveys only elicited anthropometric outcomes for children 3 and younger (not 5 and younger). However the new DHS rounds do use the 5-year age threshold for all anthropometric outcomes. Some data are also available from LSMS surveys.
Prevalence of stunting is included in UN Women and UNSC-B. MDG includes prevalence of underweight children (wasting).
3. Prevalence of anemia in women of reproductive age (15 to 49), pregnant and non-pregnant.
DHS is the primary source of information for anemia in pregnant versus non-pregnant women, and coverage is quite high (around 70 percent), with increasing coverage in the last 5 years.
Prevalence of anemia appears only in the UN Women report.
4. Maternal deaths per 100,000 live births.
Most data for developing countries are derived from models. A few observations are derived from household surveys (DHS or MICS), but large samples are needed to arrive at reliable estimates. Regular reporting of national estimates requires CRVS with cause of death. Survey data collection in most countries is not reliable or cost effective. The maternal death ratio is included in MDG, UN Women, IAEG-GS, and UNSC-B.
5. Female under-five mortality rate, and ratio of female to male under-five mortality.
Most data for developing countries come from household surveys with detailed birth histories (DHS and MICS). Annual estimates are interpolated between survey years. Regular reporting of national estimates requires civil registration of deaths with cause of death or more frequent, standardized surveys. When comparing female to male ratios, expected biological sex differences in mortality at birth should be considered.
The under-five mortality rate is included in MDG, UN Women, UNSC-B, and IAEG-GS.
6. Number of female new HIV infections per 1,000 susceptible population (by age and key populations), and ratio of female to male new HIV infections.
The most reliable data on HIV infections has come through blood samples collected during DHS surveys. However, this is a costly process, best used in places with generalized epidemics. Other data come from sentinel sites and from other administrative reporting. Measures of incidence (the number of new infections) are generally derived from estimates of prevalence (the total number of cases in a population). The MDG metadata on HIV incidence note that “…directly measuring HIV incidence is a complex process, therefore national, regional and global estimates of HIV incidence are usually produced by computer models and are based on estimates of HIV prevalence. … The estimates produced by UNAIDS/WHO are based on methods and parameters that are informed by the UNAIDS Reference Group on HIV/AIDS Estimates, Modelling and Projections, described in reports available at www.epidem.org. … The Reference Group assesses the most recent published and unpublished work drawn from research studies in different countries. It also reviews advances in the understanding of HIV epidemics and suggests methods to improve the quality and accuracy of the estimates.” HIV incidence rate is included in UNSC-B.
The HIV prevalence is included in MDG, IAEG-GS, and UN Women.
7. Adolescent birth rate per 1,000 women by age of the mother (15-19).
Data for calculating the adolescent birth rate may come from civil registration, the census, or surveys. In the case of survey data, the birth rate is generally computed from retrospective birth histories. More information on the computation of the adolescent birth rate is available in the MDG metadata.
Because the age threshold for asking questions on birth history usually excludes the below-15 age group, birthrates for younger mothers are not generally available. But household rosters of surveys like the MICS, DHS, and LSMS often identify the natural mother of each member in the household, so the share of births attributable to girls in younger age groups could be constructed.
Adolescent birth rate is included in MDG, UN Women, and IAEG-GS. The UNSC-B recommends data for age groups 10-14 and 15-19.
8. Percentage of reproductive age women (15-49) using modern contraception.
Data on contraceptive use are obtained from surveys such as DHS, MICS, or other health surveys. Data are also available in some LSMS surveys. Differences in the definition of modern contraceptive methods and differences in the population surveyed (married women or women in union or all women) and age grouping can affect comparability of the data. A related indicator, the unmet need for contraception, measures the proportion of women who desire contraceptive services but are unable to obtain them.
This indicator requires a more extensive set of survey questions and is currently only available from recent DHS surveys. The contraceptive prevalence rates is included in MDG, IAEG-GS, and UN-Women. The UNSC-B recommends an indicators of the demand satisfied with modern contraceptives.
9. Percentage of girls/young women aged 3-5 years above official age for last grade of each level of education (primary, lower secondary, and upper secondary) who have completed the level, and female to male ratio.
Household surveys of all types frequently contain information on the age and grade last attended by household members. To calculate this indicator on a cross-country comparable basis, the ISCED definitions of school stages should be employed, which may differ from national usage. Definitions of the official age would have to be correspondingly adjusted.
MDG and UN-Women include a proxy indicator for the primary completion rate (all ages). IAEG- GS also includes the primary completion rate. The UNSC-B recommends completion rates in primary, lower secondary, and upper secondary.
10. Share of women aged 15-49 who use computer and/or internet at least once a week, and every day.
A few household survey categories ask respondents about the use of computers and/or the internet, and the DHS, as well as more recent rounds of the MICS, have additional questions on the frequency of use (almost every day, sometimes, and so forth). The availability of survey data is still limited, especially in regions not covered by DHS or MICS. Some effort will be needed to initiate more general coverage.
The IAEG-GS and UN Women lists include the proportion of individuals using the internet by sex. The UNSC-B includes subscription to mobile cellular and/or fixed broad band internet (per household/100 people) and percentage of youth/adults who are computer and information literate.
11. Percentage of women aged 20-24 who were married or in a union before age 18.
This indicator is used to monitor the extent of child marriage. Both the DHS and MICS have a survey question on age at first marriage.
Age at first marriage is included in UNSC-B and IAEG-GS. UN-Women suggests reporting both marriage before 18 and marriage before 15.
12. Proportion of seats held by women in national parliament.
The share of women heads of state, government ministers, and parliamentarians is tracked systematically by the Inter- Parliamentary Union (IPU) and International IDEA. Data is updated every two years, with some of the archives going back to 1945. Data are obtained from UN country missions and embassies. IPU and UN Women publish a map every two years with this information. There is also information available on electoral quotas for women and parliamentary committees on gender from the same sources.
This indicator is included in MDG, IAEG-GS, and UN Women. UNSC-B suggests “Proportion of seats held by women in local governments.”
13. Share of women among mobile telephone owners.
Since 2013, all DHS surveys have included a question for individual men and women on whether they own a mobile phone. Other surveys typically only have individual-level questions on use of a mobile phone or aggregate ownership/use at the household level.
This indicator is included in UNSC-B. UN Women and IAEG-GS include an indicator of individuals using (not owning) a mobile telephone.
14. Share of young women who are idle – not employed + not in school + not looking for work.
Surveys such as the LFS, as well as recent rounds of the LSMS and country household welfare surveys, would allow the construction of this indicator because they have standard modules on employment that include a question on whether the respondent is looking for work, as well as a question on current schooling enrollment. This indicator is recommended as a more robust measure of young women’s attachment to the work force. Standard measures of unemployment are premised on participation in the labor force (“looking for work”) and miss those who, for whatever reasons, have been left out.
MDG, IAEG-GS, UN Women, and UNSC-B all include an indicator of the youth unemployment rate.
15. Females employed as a ratio of the working-age female population (15 to 59), and female to male ratio.
Employment rates of both men and women can differ substantially from country to country, so the ratio of female to male employment rates provides a normalized, comparative measure of women’s employment. The employment rate alone is more useful for tracking change over time and through the economic cycle in a single country. Labor force surveys are the principal source of data. LSMS surveys also include an employment module.
Employment rates of men and women are included in MDG and UNSC-B. IAEG-GS and UN Women include complex indicator focused on child care responsibilities: “Employment rate of persons aged 25-49 with a child under age 3 living in a household and with no children living in the household, by sex.”
16. Proportion of employed who are own-account (self-employed) workers by sex of worker.
This indicator is taken as a measure of vulnerable or uncertain employment. Own-account workers are defined by the ILO as workers who, working on their own account or with one or more partners, hold the type of jobs defined as a self-employment and have not engaged on a continuous basis any employees to work for them during the reference period. They are less likely to be protected by labor laws or to have recourse to unemployment insurance or other types of social protection. Labor force surveys are the principal source of data. LSMS also includes an employment module.
This indicators is included in MDG, UN Women, and IAEG-GS.
17. Women’s share of non-agricultural wage employment.
In many countries non-agricultural wage employment is the largest source of household income. In countries that are still primarily agricultural, the non-agricultural wage sector often represents the modernizing sector, providing new opportunities and higher income for women and men. Labor force surveys are the principal source of data. LSMS also includes an employment module.
This is an MDG indicator. It is not included in other lists of gender indicators.
18. Percentage of adult women with a formal nancial account or personally using a mobile money service in the past 12 months, and female to male ratio.
This is an important measure of women’s integration into the modern economy. The LSMS and Global Findex surveys from the World Bank have individual data on ownership of a formal financial accounts. Findex has an additional question on whether the respondent has used a mobile money service in the past 12 months.
A version of this indicator is included in UNSC-B:“Proportion of population with an account at a formal financial institution, by sex and age.”
19. Growth rate in adult women’s share of household earned income among the bottom 40 percent of the population, relative to that for all adult women.
This indicator can be constructed from data on individual earnings across all adult household members. Data are available from labor force and LSMS surveys. Although measurement of earnings data is problematic, affecting comparability across surveys as well, similar issues persist with overall calculation of income as well.
This is a new indicator, proposed to address issues of economic and gender equality.
20. Percentage of female children under 5 whose births have been registered with a civil authority, and the female to male ratio.
Registration of births conveys official recognition of the civil status of a child and its mother. Many countries have incomplete registration systems and some pose specific obstacles to the registration of births by mothers without authority from fathers. Complete registration of birth, marriage, divorce, and death is a fundamental responsibility of governments. Estimates of the completeness of civil registration can be derived indirectly from demographic data. Direct estimates are obtained from surveys. Both the DHS and MICS have a question in the birth history module on whether each child’s birth has been registered. In addition, some censuses include questions about the registration of births.
This indicator is included in UNSC-B.
Annex 2: Source availability
The availability of source data for the 20 indicators was assessed using the IHSN Gender Data Navigator (GDN) database. The GDN provides a searchable inventory of gender-related questions from 1,485 nationally representative survey and census questionnaires across low- and middle-income countries. Availability of the underlying source does not guarantee that a particular indicator has been constructed or published. In some cases, an indicator may be compiled from more than one source, but the methodology used in collecting or compiling the data may differ in ways that limit comparability. A few indicators (such as the share of seats in Parliament held by women) are not collected through surveys but from administrative data.
The datasets in the GDN span a range of different individual-level surveys as well as population and agricultural censuses. Individual-level surveys include Labor Force Survey (LFS), Living Standards and Measurement Study (LSMS), Multiple Indicator Cluster Survey (MICS), health surveys consisting primarily of the Demographic and Health Survey (DHS), and the 2011 Global Financial Inclusion Survey (Findex). The GDN also includes a range of surveys classified as other household welfare surveys and household budget/consumption surveys that are country-specific.
The coverage statistics included in the table were compiled for 77 IDA-eligible low- and lower-middle income countries. Coverage is computed for the 10 years, 2003-2012. Data for more recent years are still being added to the GDN.
Notes:
-Figures represent share of surveys among IDA countries in the Gender Data Navigator database (http://datanavigator.ihsn.org/); standard deviations in brackets. Country set excludes Timor Leste, which is not yet included in the GDN database.
(a) Coverage assumes a national poverty line is available for each country.
(b) Full description of indicator: Share of girls/young women aged 3-5 years above official age for last grade of each level of education (primary, lower secondary, upper secondary, and tertiary) who have completed the level, and female to male ratio. This also assumes that the official age for each grade is available in each country.
(c) Not in GDN database questionnaire.
(d) Although coverage is low in the table, DHS surveys after 2013 have individual data on ownership of a mobile phone.
Table 1. Survey coverage (2003-2012), by region, of Ready To Measure indicators among IDA-77 countries
Table 2. Survey coverage (2003-2012), by survey category, of Ready To Measure indicators among IDA-77 countries
Notes:
Figures represent share of surveys among IDA countries in the Gender Data Navigator database (http://datanavigator.ihsn.org/); standard deviations in brackets. Country set excludes Timor Leste, which is not yet included in the GDN database.
(a) Coverage assumes a national poverty line is available for each country.
(b) Full description of indicator: Share of girls/young women aged 3-5 years above official age for last grade of each level of education (primary, lower secondary, upper secondary, and tertiary) who have completed the level, and female to male ratio. This also assumes that the official age for each grade is available in each country.
(c) Not in GDN database questionnaire.
(d) Although coverage is low in the table, DHS surveys after 2013 have individual data on ownership of a mobile phone.
Annex 3: Ready to Measure indicators and the Sustainable Development Goals
The indicators proposed for the Sustainable Development Goals include fourteen for Goal 5 (“Achieve gender equality and empower all women and girls”) and numerous other indicators for other goals that can and should be disaggregated by sex. But not all of these indicators are readily available from current sources. Of the twenty “ready to measure indicators,” shown in the first column, sixteen are similar or equivalent to SDG indicators shown in the second column of Table 2.1 below. In many cases, the differences (highlighted in bold) are only in their emphasis on the calculation of female-specific shares or ratios, which will be available from sex-disaggregated data as recommended for the SDGs. Other differences in specification are noted in the third column. Thus a program to assemble the important gender indicators described here will provide a rapid start on monitoring progress toward the Sustainable Development Goals.
Table 3: Comparison of Ready to Measure (R2M) & IAEG-SDG Indicators
