The Time Use of Mothers in the United States at the Beginning of the 21st Century

Ch 6 - Concluding Remarks

121 6 Concluding Remarks Time is our most scarce resource and children our most precious. Raising children, especially young children, is inherently time intensive for parents, especially for mothers, who in every country serve as primary caregivers for most children. We refer to the child-rearing time of mothers as maternal caregiving, and throughout this book we have examined in detail the role that maternal caregiving time plays in U.S. mothers’ days. Caring for children requires trade-offs: spending more time or money on children necessarily implies spending less time and money for other purposes. Are any of these trade-offs systematic? This is one of the fundamental questions of this book. In other words, do time allocation decisions differ between mothers with younger versus older children, higher-wage mothers versus lower-wage mothers, or married mothers versus unmarried mothers? Beyond the characteristics of the mothers themselves and the characteristics of their children, we also explore the role that fathers play in mothers’ time trade-offs. Do mothers with husbands who are employed many hours per week make different time choices than mothers whose husbands work fewer hours per week? Does mothers’ time with children depend on their husbands’ time with children? Finally, we examine whether the time of day when employment occurs has implications for maternal caregiving time and the timing of that caregiving. We began with a descriptive look at mother’s time use and then turned to a statistical examination of the nature of caregiving and the ways that it differs from other time uses. All of our analyses are based on the ATUS, an annual product of the U.S. Census Bureau and the Bureau of Labor Statistics. After nearly 10 years of development, the ATUS was initiated in 2003 and the first data from this annual ongoing survey were released in January of 2005. In Chapter 2, we described this new time diary data source in detail, such as how the data are collected, the sampling, and the way time is categorized. The most important characteristics of the ATUS for our purposes are that sample 122 Connelly and Kimmel sizes are large, there is substantial demographic information available in addition to the time diaries, and only one 24-hour time period is recorded with an oversampling of weekend days. In addition, only one time diary is collected per household, although we know much about the demographic characteristics of the other members of the household. In Chapter 2 we also described the choices available in defining maternal caregiving time in the ATUS. While our statistical analyses in later chapters focused on the measure of caregiving that we refer to as primary caregiving, in Chapter 2 we explored two other potential measures: secondary caregiving and “time with children” in order to provide a fuller picture of maternal caregiving. The ATUS is not an ideal data source for a variety of reasons. First, the 24 hours of time use information is collected by recall rather than by an ongoing time diary in which activities are recorded during the particular 24-hour period. Still, the recall time is only one day instead of a week or a year, as is required in other data surveys. The strategy of a one-day recall has been well tested and judged to be a good tradeoff between overly invasive continuous surveys versus a longer recall period. A second concern is that only one day of time diary information is collected, and this single day reflects merely a random snapshot of the respondent’s time use. We cannot determine how typical the survey day is for respondents. This is less of a problem when one’s research goal is to assess average behavior but becomes highly problematic for seeking individual level causality, such as how the amount of time spent exercising affects a respondent’s weight. Our research questions fall somewhere in the middle as we try to predict time use rather than assess the value of that time use for other outcome variables. The third and most important shortcoming of the survey design, from our perspective, is that only a single time diary per household is collected. This is problematic as we are interested in the interplay between mothers’ time use and their husbands’. To compensate for this last problem, we develop a statistical methodology to estimate husbands’ time use using the time diaries of fathers. One of the ATUS’s many strengths is that time is categorized very precisely. We...