Families That Work: Policies for Reconciling Parenthood and Employment

Appendix A: Description of Cross-National Data Sets Used

Appendix A Description of CrossNational Data Sets Used CROSS-NATIONAL DATA in chapters 3 and 8 come from four main sources. Each is described here, along with information on sampling and variable definition. The data source for figures 3.1 to 3.8, 3.12, 3.13, 8.2, 8.3, and 8.8 is the Luxembourg Income Study (LIS) (for general information, see: www.lisproject.org). (Detailed information on the original surveys, including sample sizes, is available at: www.lisproject.org/techdoc/surveys.htm.) The Multinational Time Use Study (MTUS) is the source for figures 3.9, 3.10, and 8.5 (for general information, see: www.iser.essex.ac.uk/mtus). (Detailed information on the original surveys, including sample sizes, is available at: www.iser.essex.ac.uk/mtus/structure.php notes.) Eurostat’s European Labor Force Surveys (LFS) are the data source for the European countries in figures 3.11 and 8.4 (for general information, see: europa.eu.int/comm/eurostat). Data on the United States in figures 3.11 and 8.4 are from the 1997 Current Population Survey (CPS); the U.S. results were provided by Harriet Presser. The International Social Survey Programme (ISSP) is the data source for figure 3.19 (available at: www.issp.org). (Detailed information on the surveys and modules is available at: www. gesis.org/en/data—service/issp/index.htm.) LUXEMBOURG INCOME STUDY The Luxembourg Income Study is a public-access archive of microdata sets from approximately twenty-five industrialized countries , with data available for up to five points in time, depending 306 Families That Work on the country. The LIS data include demographic, income, and labor market indicators; data are available at the person and household level, and records can be linked. Most of our comparisons based on the LIS data include all twelve comparison countries and cover the time period from 1992 to 1997. An exception are analyses related to working time; weekly hours are available for only nine of these countries (and including Finland necessitated using data on weekly hours in Finland from 1991, as the Finnish data exclude hours worked in 1995). A second exception is the analysis of the part-time wage penalty, in which we were able to locate data for only five of the twelve countries. In addition, though later Danish data are available through the LIS, we used the 1992 data set because of missing or problematic variables in later data sets. In the LIS-based analyses, we selected persons aged twentyfive to fifty or couples in which either partner was aged twentyfive to fifty. Couples include both married and cohabiting persons, wherever possible. When reporting employment rates, we exclude workers in the agricultural and military sectors; when reporting weekly hours, we further exclude the self-employed. In the German data set, we include the western Lander (states) only, and in the U.K. data set we include only England, Scotland, and Wales—in both cases to raise the comparability with both available policy data and the ISSP data. Employment status (yes or no) is coded from labor-force-status variables generally referring to the prior week (where “no” includes those out of the labor force and the unemployed), and weekly hours refer to hours usually worked each week. Labormarket earnings are gross (pretax) earnings with one exception— France—where they are net (posttax). Poverty is defined as having income below 50 percent of the national median posttax and posttransfer income, with an adjustment for family size. Details on coding of specific variables are available from the first author (Janet Gornick) on request. MULTINATIONAL TIME USE STUDY The Multinational Time Use Study (MTUS) compiles comparable time-use-diary data from a number of countries. The MTUS data Cross-National Data Sets 307 were available for only eight of our twelve comparison countries; the time period used is 1985 to 1992. In the MTUS-based analyses, the sample is restricted to married (or cohabiting, where available) couples in which the adults are between the ages of twenty-five and fifty with at least one child under the age of five living in the household. Measures of unpaid work in the home include time spent in routine housework (cooking, cleaning up, and housework), in nonroutine housework (other domestic work, gardening, shopping, domestic travel), and care for children. Only primary activities are recorded. Time is recorded daily and averaged (in minutes...