Users, uses and production of gender statistics: an overview
What are gender statistics
Users and uses of gender statistics
Production of gender statistics
A guide to the present manual
Bringing gender issues into statistics
Outcome of formal education
Non-formal adult education and training
Scientific and technological knowledge
Labour force participation
Access to productve resources in agriculture
Reconciliation of work and family life
Household-level income / consumption poverty
Inequality in intrahousehold allocation of resources
Economic autonomy of women
Environmental aspects with gender-differentiated impacts
Involvement of women and men in preserving the environment
Power and decision-making
Politics and governance
The private sector
Population, households and families
Demographic composition of population
Formation and dissolution of unions
Fertility and contraceptive use
Health and nutrition of children
Mortality and causes of death
HIV and AIDS
Health risk factors related to life style
Migration, displaced persons and refugees
Refugees and internally displaced persons
Violence against women
Physical and sexual violence against women
Female genital mutilation
Integrating a gender perspective into data collection
Population and housing census
Agricultural censuses and surveys
Labour force survey
Time use surveys
Surveys on violence against women
Analysis and presentation of gender statistics
Descriptive analysis of data
Presentation of gender statistics in graphs
Presentation of gender statistics in tables
Table of contents
Abbreviations and acronyms
Glossary of terms
Read in Order
Create a new Page
Time use surveys
Modified on 2015/05/26 07:48
Read in Order
Chapter 3 - Integrating a gender perspective into data collection
This section of the manual draws extensively on the United Nations publication
Guide to Producing Statistics on Time Use: Measuring Paid and Unpaid Work
(2005). Readers are strongly encouraged to consult the above-mentioned publication, as more detail as well as additional information are covered at length.
Uses of time-use data for gender statistics
Time-use data show how individuals allocate their time to specific activities over a specified period - typically over the 24 hours of a day and over the seven days of a week (United Nations, 2005). These data are generally obtained through two types of survey instruments: 24-hour time diaries and stylized analogues of these diaries. In time diaries, the respondents report all activities undertaken during a prescribed period of time and the beginning and ending time for each activity. In stylized versions of diaries, respondents are asked to recall the amount of time allocated to certain activities during a specified period of time. Compared to stylized questions, diaries can be a more reliable tool for measuring time use and therefore a more reliable tool in obtaining gender statistics; however, the resources required and the burden on the respondents are considerably higher.
Time-use data are the basis for obtaining gender statistics related to several topics: time allocation patterns; unpaid work; participation in all forms of paid work; working time, work locations and the scheduling of economic activities; work-family balance; the investment of time in education and health; welfare and quality of life; and intrahousehold inequality.
Time-allocation patterns. Time-use data can show differences between women and men in time-allocation patterns (types of activities and their schedule during a specified period of time), reflecting differences in roles and expectations with regard to family, domestic life and participation in work and social activities outside the home. Women, for example, tend to spend more time than men taking care of children and less time than men working for payment outside of their homes. The schedule of various episodes of activities may also be different for women and for men. For example, men may be more likely than women to spend time on paid work activities during the night; or, time spent with children may be more concentrated for men in the evening hours and during the week-end. Gender differences extend beyond the traditional distribution of roles in paid and unpaid work. Different patterns of leisure activities may be described for women and for men. In some societies, for example, men may spend more time than women doing sports and fitness activities.
Unpaid work. Time-use data are essential in estimating the participation of women and men in unpaid work (activities unaccompanied by remuneration) and the value of this unpaid work for the economy. Some types of unpaid work are not covered by conventional labour statistics or the System of National Accounts. Conventional labour statistics are limited to activities that contribute to the production of goods and services, as defined by the System of National Accounts, and cover mainly market activities and some unpaid non-market activities. Unpaid work referring to own-account production of services is outside the general boundary of the System of National Accounts and therefore not covered at all in conventional labour force statistics. Examples of unpaid work include cleaning dwellings, performing small repairs, preparing and serving meals, caring for and instructing children, caring for other persons in the household and certain types of volunteer community services. Data on time use for those types of activity are typically obtained in time-use surveys or time-use modules attached to labour force surveys, living standard surveys or other multitopic surveys. The data obtained can be used to estimate household production in satellite accounts that extend measurement of gross domestic product (GDP) to include non-System of National Accounts production, making the national accounts more complete and comparable across countries. At the same time, time-use data on unpaid work is crucial to making the contribution of women to the economy and society more visible. Women, more often than men, tend to be involved and spend a great amount of time in unpaid work in the home and community. When only cash transactions are taken into account in the measurement of economic production, a large portion of women’s work remains unaccounted for.
Participation in all forms of work. Time-use data have an important role in improving estimates of employment and labour force participation through more extensive capturing of the participation in non-market activities that are within the general production boundary of the System of National Accounts. These activities refer to the production of goods for own consumption, such as agricultural work, fishing, hunting, cutting firewood, carrying water, threshing and milling grain, making butter and cheese and slaughtering livestock. Such activities are taken into account by current concepts of labour force and employment and should be covered by conventional labour statistics. As a result of bias in data collection, however, these activities are often underreported in labour force surveys or censuses. Women in particular tend to have their participation in labour force underreported, because they tend to be more often involved in non-market economic activities and because it is often assumed (by interviewers or respondents themselves) that women’s activities are limited to the domestic area. Time-use surveys are more suited for capturing the involvement of women and men in atypical cases of non-market activities that should be considered within the general production boundary of the System of National Accounts, and for obtaining a measure of the amount of time allocated to those activities.
Working time, work locations and the scheduling of economic activities. Characteristics of work such as working time, work locations and the scheduling of economic activities are often different for women than for men. Time-use data can be used to improve the measurement of time spent for economic activities and to better identify work locations and the scheduling of activities during the week and within 24 hours. Time-use surveys may be able to provide better estimates of working time than labour force surveys, especially when the distinction between periods of work and non-work may be unclear or when such periods are frequently interchanged. They may also provide better estimates of when work is carried out, and at what locations. Work is increasingly undertaken in non-traditional places. A good portion of paid work may be performed at home or while commuting. Contextual information (i.e., for whom and for what purpose is the activity done) is important in order to identify activities outside the workplace that are actually performed for an institution, even if they appear to be personal. Reporting work-related activities may also help to identify, non-paid activities that are performed for work. For example, self-employed persons may carry out various activities that are important for their business but are not formally remunerated, such as socializing.
Work-family balance. Time-use data can provide great insights into the gender specifics of the interaction of work and family life and the relationship between the labour force participation of various female and male household members and their involvement in domestic care activities. Time-use data can also show how the gender division of labour is changing with the change in the balance between market and non-market activities. This is especially the case when there is supplementary information available regarding local infrastructure, domestic appliances and the consumption of market services that substitute for households’ own labour (maids, childcare centers and nursing care, for example).
Investment of time in education and health. Time-use data can contribute to understanding gender differences in investment in education and health. Data on children’s investment in education may refer to time use spent on activities such as being in school and doing homework or other learning activities, as well as on the balance between time spent on learning activities and time spent in employment or doing housework. Data on time use can also be used to better understand gender differences related to physical access to school when girls and boys do not use the same means of transportation or do not attend the same schools. Furthermore, data regarding the use of technology such as the Internet, computers or telephones in performing various activities can be used to assess gender gaps in access to communication and technology.
Time-use data can be used to understand the relationship between gender and health. Commonly, time-use data cover activities of care for the sick and disabled, which are more often the responsibility of women than of men. Time-use data may also cover transportation to health facilities and waiting time to obtain a consultation, which may show gender differences in accessing health services. Lastly, some data needed to assess the time lost to ill health, which are often used as a measure of health status for individuals and as a non-consumption indicator of poverty, can also be obtained in time-use surveys or multipurpose surveys collecting data on health and time use.
Welfare and quality of life. Time-use data on leisure and the psychological well-being experienced while performing various types of activities can be used as measures of welfare and quality of life for women and men. In certain contexts, the shortage of leisure can be an indicator of poverty. Poor people have to spend most of their time producing the income required for basic needs and do not often have time for leisure. Leisure may include social interactions, relaxation, cultural activities, solitude, physical exercise or participation in sports or games; however, the definition of an activity as leisure is cultural and varies from person to person. Cooking is often given as an example of an activity that may be a chore for one person, but leisure for another person. Women are usually responsible for cooking, and this activity is often identified as a household chore; however, cooking, for example, may be performed once in a while by the husband, as a hobby, and therefore it may be identified as leisure. Shopping is another example of an activity that can be defined as household chore or individual leisure.
Variables that measure psychological well-being while performing each activity (for example, whether people experience negative emotions such as tension or stress or positive feelings such as enjoyment or content) can also be used to differentiate between activities that are performed as duty and activities that are performed as hobby, and to give an indication of the quality of life. Women are most often in charge of taking care of their children, for example. A woman may experience positive feelings while spending time with her children when the availability of childcare services prevents the time invested from becoming a burden. When care services are not available and the woman is the only care provider, however, negative emotions and lower quality of life may be associated with some of the activities related to the care of children.
Intrahousehold inequality. Data related to the amount of time spent by all household members on household chores can be used as an indicator of household and individual living standards and help redefine poverty in terms of the lack of time resources. At the household level, for example, “household time overhead” is defined as the minimum number of hours that a household must spend on the basic chores vital to the survival of the family, such as the preparation of meals, washing clothes, cleaning the house and time spent fetching water and firewood for cooking. A household with low household time overhead is better off than a household with a high time overhead. How the burden of work is distributed among women and men of the household is an important contribution by time-use data to the understanding of intrahousehold inequality. The household maintenance tasks are not distributed evenly among household members and have a different impact on women and men. As a result, women are, more often than men, “time poor”. Furthermore, data on time use can show how the gender allocation of time within the household is changed when access to public services, infrastructure or domestic appliances is provided. They can also show whether the household time overhead is covered by girls or boys when adults, especially women, spend more time on market activities, and whether there are gender differences in the impact on time devoted to schooling and learning activities.
Avoiding gender bias in data collection
Several aspects specific to time-use surveys are important in avoiding gender bias in data collection: the classification of activities covered; the recording of contextual information; the recording of simultaneous activities; the coverage of relevant individual and household characteristics; and sample coverage.
Classification of activities
Time-use classifications should be detailed enough to identify separately the activities mainly undertaken by women or mainly undertaken by men. It is important that activities with great gender differences in time use be collected in as detailed a manner as possible and later, at the stage of the coding, analysis or presentation of data, they are not collapsed into larger categories where the gender differences disappear. Also, when using light diaries or stylized analogues of time diaries, the predefined list of activities of interest should capture the specificity of the type of activities carried out by women and men, and the different amount of time allocated to each. It is important that the data collected be as detailed as possible. For example, instead of one category of care for children, adults or the elderly, more categories should be defined, such as care of children, care of ill adults, care of the elderly and care of the disabled. Fetching water and fetching firewood should also be included as separate activities among other types of non-market work.
Time-use classifications should allow for a distinction between market work, non-market work, domestic activities and volunteer work and enable the provision of data that can be linked to official statistics emanating from the System of National Accounts and labour statistics frameworks. This is especially critical in situations in which time-use data are used to measure unpaid work in satellite accounts that extend the measurement of GDP to include non-System of National Accounts production. Furthermore, the distinctions between different types of work are particularly important in understanding the specific contribution of women and men to the economy and society.
For example, the International Classification of Activities for Time-Use Statistics uses the System of National Accounts as a basic framework and distinguishes between productive activities within the System of National Accounts, productive activities outside the System of National Accounts, and personal activities. Also useful for understanding gender differences are the categories of unpaid non-System of National Accounts activities included in the classification. For instance, three of the major groupings of the classification are “unpaid domestic services for own final use within households”, “unpaid caregiving services to household members” and “community services and help to other households”.
Recording of contextual information
The separation between different types of work and between work and leisure is possible only when additional contextual information is collected in the time-use surveys. The context in which activities take place include the location of an activity, the other people present, the person or institution for whom the activity was done, the purpose of the activity and any remuneration that may have been received for the activity. Contextual information is crucial to coding and classifying activities reported in the time-use surveys. For example, information on the persons or institutions for whom a particular activity was being carried out, and whether payment was involved, would be needed to identify volunteer work, unpaid work within the household and unpaid work outside of the household.
Other contextual information can be useful from a gender perspective. For instance, subjective information on well-being experienced while performing an activity may be used in constructing measures of quality of life for women and men. Also, information on the ownership of assets and durable goods such as domestic appliances, and information on types of fuels used in the household and types of water sources can be used for valuing unpaid work and the construction of satellite accounts.
Recording of simultaneous activities
Time use for specific types of activities, often related to unpaid work and often performed by women, can be adequately identified only when simultaneous activities are recorded. An activity may be carried out in parallel with one or more other activities over an interval of time. For example, a woman may take care of her children while cooking or while doing the laundry. When estimates of time use are based only on primary activities, many activities, such as caring for children, older persons, ill or disabled persons, for example, are clearly underestimated. These “missing” activities would typically be reported as secondary or simultaneous activities.
A large proportion of secondary activities tend to be underreported. For example, respondents may perceive domestic or personal care activities as not important. Probing questions can bring out unreported simultaneous activities. When multiple activities are recorded in a questionnaire for the secondary activity and only one activity can be recorded in the database, priority should be given to activities such as unpaid domestic work, childcare, care of the elderly and care of the sick and disabled.
The ability to collect data on simultaneous activities depends on the methods of data collection used. For example, it is difficult to record simultaneous activities through a telephone interview. Underreporting of secondary activities may be more likely to appear when using activity lists and stylized questions that constrain people to summarize their activities into a total of 24 hours.
Coverage of relevant individual and household characteristics
Gender differences in time allocation for specific activities tend to vary in different groups of population and at different life stages. Individual characteristics commonly recorded in time-use surveys refer to sex, age, marital status, the presence of children, education and labour force status. This information will allow time-use data to be disaggregated at the level of particular groups of women and men. When time-use data is collected for only one person in the household, and the questionnaire needs to be kept to a minimum, it is important for the demographic, social and economic characteristics of the partner of the respondent to be recorded. When data on time-use is collected for all relevant household members and more extensive background characteristics can be recorded, however, individual characteristics should be captured for all household members. In addition, information on housing characteristics or other measures of wealth of the household should be collected. Such information will enable the identification of patterns of intrahousehold allocation of time resources according to various types of living arrangements and wealth group of households.
Samples used in time-use surveys should cover all relevant groups of population and all seasons. In particular, children and older persons should not be excluded by the survey sample. It is also important for the sample of days included in the survey to cover all seasons relevant to agricultural or other weather-dependent activities.
Data on time use for multiple persons from the same household are the basis for understanding intrahousehold allocation of time and resources. A full accounting of time use during the past 24 hours for all eligible household members is necessary in order to study how housework and childcare are allocated among household members, and how a change in one person’s labour activities affects the use of time by other household members.
Bittman, Michael (2000). Issues in the design of time-use surveys for collecting data on paid and unpaid work. Paper presented at the Expert Group Meeting on Methods for Conducting Time-use Surveys. New York, 23-27 October.
Harmonized European Time-Use Surveys: 2008 Guidelines.
Harvey, Andrew S., and Maria Elena Taylor (2000). Time use. In
Designing Household Survey Questionnaires for Developing Countries: Lessons from 15 years of the Living Standards Measurement Study.
Margaret Grosh and Paul Glewwe, eds. Washington, D.C.: World Bank.
Hoffmann, Eivind, and Adriana Mata (2000). Statistics on working time arrangements: issues and the role of time use surveys. Paper presented at the Expert Group Meeting on Methods for Conducting Time-use Surveys. New York, 23-27 October.
Grosh, Margaret, and Paul Glewwe, eds. (2000).
Designing Household Survey Questionnaires for Developing Countries: Lessons from 15 years of the Living Standards Measurement Study.
Washington, D.C.: World Bank.
United Nations (2005).
Guide to Producing Statistics on Time Use: Measuring Paid and Unpaid Work.
Series F, No. 93. Sales No. E.04.XVII.7.
___, Economic Commission for Europe, and World Bank Institute (2010). Developing gender statistics: a practical tool. Geneva. ECE/CES/8.
version 126.96.36.1990. Some of the icons created by