Our data come from the Multinational Time Use Study (MTUS) (www.timeuse.org/mtus/), an ex-post harmonized cross-time, cross-national comparative time-use database, constructed from national random-sampled time-diary studies with detailed measures of time use. The MTUS aggregates daily ‘primary activity’ in 40 time use categories (and an additional category for missing time), with approximately 30 standardized demographic variables. All the surveys provide weights designed to ensure that the surveys are nationally representative. The following table shows the countries and surveys we use in our analysis, and gives the total (unweighted) numbers of diary evidence available in for each country and survey.
Table A.1 Description of the Time Use Surveys
Download Excel sheet
The sample is restricted to include non-retired non-student individuals between ages 18-65 who had time diaries that summed to a complete day ( i.e., 1140 minutes) and reported information on age, education, and the presence of a child in the household.
To test the different explanations to the trends in the education gradient, we use additional information from micro-level and secondary macro-level data sources. In particular, we use microdata from the International Crime Victims Survey (ICVS) (urn:nbn:nl:ui:13-wx0-h0o) (van Kesteren, J.N. 1989; 1992; 1996; 2000; 2005) to analyze changes in parents’ perceptions of safety over time. We also use the information collected by three waves of the World Values Survey (WVS, 1994-1999, 1999-2004, 2005-2007) (http://www.wvsevsdb.com/wvs/WVSData.jsp) on which qualities do parents prioritize when upbringing children. Statistical information from secondary sources is also employed. We use gross pre-primary enrolment rates and gross tertiary education enrolment rates from the World Bank Data (World Bank 2011) (http://databank.worldbank.org/ ddp/home.do?queryId=189), parental leave regulations from the Comparative Family Policy Database (Gauthier 2011) (www.demogr.mpg.de), and teaching and academic staff employment in pre-primary education from the OECD Education Database (OECD 2012) (http://stats.oecd.org/#).
The following table summarizes the characteristics of these samples, number of observations, survey years for the different countries, and main variables used from each data set.
Table A.2 Description of Additional Datasets
Download Excel sheet