From the paper presentations and discussions during the workshop, several overarching themes emerged. This section summarizes these common themes, lacing together key threads of discussions from each of the workshop sessions.
Data on time use are important sources of information, and the lack of national time-use data is a critical gap in the federal statistical system. Time-use data produced on a regular and on-going basis can advance knowledge of the well-being of the U.S. population and can be used to inform public policy. Some examples of how time-use data can be used that were discussed in detail at the workshop include: better measures of labor inputs for productivity statistics; improvements in the coverage of national income and product accounts; understanding the changing nature of child care and elder care; understanding the effects of welfare reform; additional understanding of the role of retired persons in the nation; and better understanding of the “time crunch” felt by many people.
Time-use data can also be used to further understanding of household behavior, including the allocation of time and goods among household members and subjective feelings and satisfaction levels associated with time spent in different activities. Time-use data are also important for making international comparisons. Improved coverage in national income and product accounts that include measures of nonmarket production can enhance our
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 58
Time-Use Measurement and Research: Report of a Workshop 7 Summary From the paper presentations and discussions during the workshop, several overarching themes emerged. This section summarizes these common themes, lacing together key threads of discussions from each of the workshop sessions. IMPORTANCE OF U.S. TIME-USE DATA Data on time use are important sources of information, and the lack of national time-use data is a critical gap in the federal statistical system. Time-use data produced on a regular and on-going basis can advance knowledge of the well-being of the U.S. population and can be used to inform public policy. Some examples of how time-use data can be used that were discussed in detail at the workshop include: better measures of labor inputs for productivity statistics; improvements in the coverage of national income and product accounts; understanding the changing nature of child care and elder care; understanding the effects of welfare reform; additional understanding of the role of retired persons in the nation; and better understanding of the “time crunch” felt by many people. Time-use data can also be used to further understanding of household behavior, including the allocation of time and goods among household members and subjective feelings and satisfaction levels associated with time spent in different activities. Time-use data are also important for making international comparisons. Improved coverage in national income and product accounts that include measures of nonmarket production can enhance our
OCR for page 58
Time-Use Measurement and Research: Report of a Workshop knowledge of differences in output across countries and improve our ability to compare the output and income of the United States with those of other high-income countries and countries with developing economies. Time-use data can also be used to help understand cultural and social differences across countries. Efforts to collect data on time use in other countries are more advanced than those in the United States. Australia and Canada both have regular and comprehensive surveys for collecting time-use data on a national basis. A harmonized European time-use survey that will be conducted in almost 20 countries is also moving forward through Eurostat. The United States does not currently collect regular and comprehensive time-use data on the American population. PROPOSED BLS TIME-USE SURVEY The Bureau of Labor Statistics has issued a report on the feasibility of conducting a time-use survey of the U.S. population. The report proposes to collect time diaries from a sample of the adult population of the United States from members of the outgoing rotation of the Current Population Survey. Overall, workshop participants acknowledged the need for and value of a national time-use survey and greeted the prospect of a future BLS time-use survey with enthusiasm. The proposed time-use survey can go a long way towards furthering understanding of many of the policy and behavioral issues discussed in this summary. A common theme emerging from the workshop discussion is that the BLS proposal is, on the whole, timely and carefully designed and ready to be taken to the next stage of development and refinement as a prelude to full deployment. Economic and Demographic Characteristics A number of design features of time-use surveys were highlighted at the workshop. To achieve a better understanding of household time allocation, nonmarket household production, and the effects of public policy on time use, many workshop participants emphasized the importance of collecting the fullest possible array of individual and household-level economic and demographic variables as possible. Such variables include, but are by no means limited to, age, race, gender, household structure and size, age and number of children, education levels of household members, income and wealth, labor force status, occupation, and wage rates. The CPS already collects data on many of these characteristics and the proposed BLS time-use survey plans to update these data. In addition to the regular CPS-collected data, workshop participants stressed the importance of collecting the widest possible array of background characteristics. Collecting the broadest array of
OCR for page 58
Time-Use Measurement and Research: Report of a Workshop covariates as possible is especially important for research uses of the data. The range of topics that can be researched with the data will expand as the range of covariates collected expands. Multiple Diaries from the Same Respondent Time use by a household can vary greatly from day to day and from season to season, because of work schedules, schooling schedules for households with children or educators, or simply because of atypical events that occur from time to time. Because of variation across days of the week, many workshop participants urged the BLS to collect data for multiple days of the week for each respondent to better capture the variability in time use across different days of the week. Workshop participants also discussed the significant benefits of collecting longitudinal time-use diaries, that is, collecting diaries from each respondent at different times of the year. Longitudinal diaries could be collected to understand variation in time-use activities across seasons and they can be used to assist researchers in modeling changes in time-use behavior between two points in time. For example, if a person becomes employed between the two dates for which the time diaries are collected, one could examine changes in time allocated to nonmarket household production and to leisure activities for people who experience such a change in employment. Interviewing respondents multiple times however, does present survey design and cost considerations, especially since the outgoing panel of the CPS, which is comprised of respondents who have already been through several rounds of surveys, will be used to develop the time-use sample. Careful consideration of such design issues will need to be made. There are precedents in collecting multiple diaries from survey respondents, from which lessons can be drawn: the Michigan studies collected diaries at four different times over the course of the year and Australia’s studies collect data for two consecutive days from each sample member. Diaries from Multiple Household Members Many policy and behavioral questions about household time use involve the interaction of time use among family members. For example, if a wife receives a wage increase, the husband’s time spent in market and nonmarket activities may change since the relative price of the husband’s time to the wife’s time has now decreased. Another interesting question is how the Earned Income Tax Credit affects both husbands’ and wives’ time spent in market work. Interactions between children’s time use and parents’ time use are similarly important in understanding household behavior. To address these questions, data for multiple members of a household would need to be col-
OCR for page 58
Time-Use Measurement and Research: Report of a Workshop lected. Workshop participants acknowledged that there are budget and sample design issues to consider in collecting data from multiple household members, but that diaries from multiple household members would be valuable in addressing important policy and behavior questions. Data for International Comparisons A key reason for conducting a time-use survey is to measure time use in nonmarket activities so that satellite accounts to the national income and product accounts can be produced. Household input and output tables similar to those produced by Landefeld and McCulla (1999) have now been produced in five other countries (Australia, Canada, Finland, Sweden, and Norway). No doubt, when the harmonized Eurostat time-use study is completed, more countries will produce these tables. In order to make comparable estimates of nonmarket household production across nations, and to make cross-country comparisons in time-use behaviors, it is important for the proposed U.S. time-use data collection to be as comparable as possible to other nations’ studies. This need includes comparability in the classification of activities, for which there are several existing coding schemes for activities, including the United Nations International Trial Classification System for time use across different countries (see Horrigan et al. 1999 for a summary of the major classification schemes). With any standard that is implemented, special care will need to be taken to understand how question wording, examples used, and interviewer training differs across countries’ surveys and how responses might reflect these differences. Despite the difficulties in standardizing data across countries, many workshop participants emphasized the value of time-use data for making cross-country comparisons and for informing policy. FUTURE DATA COLLECTION AND METHODOLOGICAL RESEARCH No single survey will be able to collect all the data that can contribute to the policy questions that time-use data could address. For example, Katharine Abraham noted that the sample size proposed in the BLS survey will be too small to study the welfare population or those with disabilities. Also, to better understand household production, information on the technology available to the household and on the goods and services the household purchases would be needed. The CPS does not contain this information, and the BLS time-use survey will most likely not be able to collect it regularly because of scarce resources, although it may be possible to conduct a module to the CPS to collect such information or information on other specific topics. Because no single time-use survey is going to be able to include all the information needed for policy-related research, to study specific populations
OCR for page 58
Time-Use Measurement and Research: Report of a Workshop (such as welfare recipients), or to study specific themes, time-use data could be collected as part of other surveys. Strategies would need to be developed for incorporating time-use studies and questions as part of existing survey programs or in new surveys targeted to the specific populations or topics of interest. For example, time-use questions and time diaries that parents filled out for their children were collected in 1997 as part of the Panel Study of Income Dynamics to better understand children’s time use. Other potential examples mentioned at the workshop were to add a time-use diary to the Health and Retirement Survey for information on retirees and the elderly and to use the experiential sampling method to better understand time use in the workplace. New, targeted surveys of time use, like the California Activity Pattern Survey (described in Chapter 4) could also be developed. The feasibility of such studies will need to be further explored. There are also several methodological considerations for time-use surveys that need further study, such as: designing stylized questions to obtain better estimates of time use in specific activities, assessing the quality of and exploring methods to improve recall on diary surveys and on stylized questions, using the experiential sampling method on a wider scale to collect time-use data, surveying multiple household members, and collecting diaries on multiple days for each sample member. In order to investigate some of the time-use topics of interest raised in Chapter 2, special methods may need to be considered and tested. For example, to study the time use of those with disabilities, questionnaire and time diary content considerations (perhaps asking if anyone helped the respondent with the activity) may need to be considered. As discussed above, a study that uses the experiential sampling method may be the most feasible way to collect data on time use in a market work setting. Workshop participants also stressed the importance of using developments in other diary surveys to improve time diary methods, for example, travel diaries, expense diaries, or food consumption diaries. To further promote time-use surveys, many participants said that mechanisms for encouraging methodological research are needed. Time-use data can be applied to a broad range of social and economic behavioral and policy topics. Improving methods for collecting such data promises to be a rich area for research for the statistical community.