Sources of gender statistics Angela Me UNECE Statistics

Sources of gender statistics Angela Me UNECE Statistics Division 1

What Statistical Sources are relevant for Gender statistics? All sources where data on individuals are collected are in principle relevant for gender statistics Including sources used preliminary for other purposes (economic for example) UNECE Statistical Division 2

What Statistical Sources are relevant for Gender statistics? n Census n Surveys n Administrative Records n Published data n Other sources (surveillance systems, associations) UNECE Statistical Division 3

Population and Housing Census Complete Count of the people and the housing units and the collection of a number of their characteristics in a territory of a country UNECE Statistical Division 4

Population and Housing Census n n Identify each member of the population Collect certain basic data about them n n age, sex, education, employment, etc. Modules to collect data on specific topics may be added Normally about every 10 years Provide the benchmarking for population, and sampling frames UNECE Statistical Division 5

Population and Housing Census n Advantage n n Excellent coverage Disadvantage n n n May be inaccurate due to infrequency Limited data collected Lag before data produced UNECE Statistical Division 6

Population and Housing Census Data for Gender analysis n n n n Population (counting, age, marital status, family composition) Employment (occupations, status in employment, industry) Unemployment Activity Status Education (attendance, level, literacy) Migration (foreign-born, foreigners, refugees, internal migration) Living conditions (housing characteristics) UNECE Statistical Division 7

Population and Housing Census Data for Gender analysis n The census is one of the pillars in a national statistical system and in addition to the information that provides it represents the basis for the other data collection on individuals UNECE Statistical Division 8

Sample Surveys Sampling techniques are used to select a small proportion of the population that is believed to be representative of the population at large. A survey is then conducted using this sample population to gain estimates for the total population. The sample design and the sample size determine the quality and the representativeness of the data UNECE Statistical Division 9

The “Miracle” of the selected sample: how a limited number of people can provide data for the whole population? How many people is sufficient to sample in order to know the sex composition of the population? Any sample of 1 will give the sex composition UNECE Statistical Division 10

The “Miracle” of the selected sample How many people is sufficient to sample in order to know the sex composition of the population? A “good” sample of 2 will give the sex composition UNECE Statistical Division 11

The “Miracle” of the selected sample The sample of 3 could cover the all composition BUT A good sample of 4 gives the right proportions UNECE Statistical Division 12

Sample n A good sample depends on: Size (not percentage over the total population) Design UNECE Statistical Division 13

Sample Surveys n n Good vehicle for collecting data from a subset of the population Subset: n n Save money and resources Reduce time to collect data Reduce time to analyze Topics can be investigated in more details UNECE Statistical Division 14

Sample Surveys As long as The sample is properly selected: n Updated frame n Proper size (depending from the topic under investigation –its variability- and NOT on the percentage over the population) n Proper design Usually a sample of households is drawn and data collected for each member of the household UNECE Statistical Division 15

Sample Surveys Limitations n n n Sample size determines if results are generalizable to entire population (larger samples and better designs can produce better data and reliable for subpopulations/geographical localities) Data on small sub-populations may not be reliable Information on small geographic areas may not be available UNECE Statistical Division 16

Sample Surveys Data for Gender analysis n Usually household surveys focus on socio-economic issues. n Surveys that should be carried out REGULALRY: n n Income and Expenditures (Household Budget Survey) Employment (Labour Force Surveys) Health Examples of ad-hoc Surveys/Modules: n n Time-Use Survey Violence against women Employment (occupations, status in employment, industry) UNECE Statistical Division 17

International Programmes of Sample Surveys n Multi-indicator Cluster Survey (MICS) n n n n Children malnutrition Education (children and women – attendance) Reproductive health of women Children health Health status of children and women (HIV) Disability of children Infant and child mortality Demographic Health Survey (DHS) n n Same as above Violence against women UNECE Statistical Division 18

International Programmes of Sample Surveys n Living Standards Measurement Surveys (LSMS) n n n Household income and expenditures Health Education Employment Accessibility to services ……. . UNECE Statistical Division 19

Administrative (or routine) data sources n n Generated as a byproduct of events and processes and data collected by a variety of organizations (hospitals, schools, …) Primary purpose is management of processes Event triggers data production Summary and/or dissemination occurs later (but usually within one or two years) UNECE Statistical Division 20

Administrative (or routine) data sources n Examples relevant for gender analysis Vital registration n n Health system n n Diseases, services provided Education system n n Births, deaths, marriages Enrollment, teachers Employment Business registration Voting registers UNECE Statistical Division 21

Administrative (or routine) data sources • • • Advantages Less expensive than surveys and censuses Relatively up to date (usually available within one to two years after event) If properly maintained, full coverage Often produced by agencies who are stakeholders in the policy process, e. g. , health providers, schools, industry bodies, so incentive to participate Routine collection of sub-population identifiers UNECE Statistical Division 22

Administrative (or routine) data sources Disadvantages n n n n Require large efforts by Governments and People to properly maintain them Coverage may be insufficient or biased Limited set of information collected Some data may depend upon uptake of services May measure service provision rather than demand, and uptake rather than impact Numbers may be inflated in some areas Primary purpose is NOT data collection UNECE Statistical Division 23

Census and Surveys: initiated by the statistical authorities UNECE Statistical Division 24

Administrative Records: initiated by the individual UNECE Statistical Division 25

Data Sources Data for Gender analysis n For data where women and men do not have a benefit or do not see the advantage of reporting the event Household Surveys or Census are better sources UNECE Statistical Division 26

Demographic Surveillance Systems Data for Gender analysis n n Longitudinal monitoring of sentinel populations Can provide detailed information on the sentinel population but not representative of the population UNECE Statistical Division 27

Other Sources Data for Gender analysis n Professional Organizations n n n Business Journalists Lawyers UNECE Statistical Division 28

Data sources compared Characteristic Inclusion criterion Coverage Admin Survey Census All ‘noticed’ events Designated units All units Variable, depending upon system % coverage specified ~100% coverage May be biased Designed to minimize bias Lack of coverage may lead to some bias Gender Bias UNECE Statistical Division 29

Data sources compared Characteristic Admin Survey Census Cost Cheap Medium Expensive Time Ongoing, + 1 -2 years for reporting 3 -5 years + 1 year for reporting 10 years + 1 year for reporting V good, but limited info, and problem if coverage poor Good, but only for medium to long term Good for long term and as input with other data Potential for Gender analysis UNECE Statistical Division 30

Messages Look for all potential Sources Use the sources at the best for gender analysis understanding their strengthens and limitations UNECE Statistical Division 31