Abstract
School-age children need 10–11 h of sleep per night. It has been well-documented that lack of sleep leads to diminished cognitive performance and that people who sleep less are more likely to be overweight or obese. I use data from the American Time Use Survey to examine two factors that can potentially influence the amount of time children sleep: school and maternal employment. I find that school-age children sleep less when school is in session than during the summer, and that they get less sleep on school nights than on non-school nights. Children go to bed about 38 min earlier on school nights, but wake up about 72 min earlier on school days. This translates into about 34 min less sleep on school nights compared with non-school nights, and implies that these children have a cumulative sleep deficit of over two-and-a-half hours by the time they arrive at school Friday morning. In addition to the lost sleep time, the earlier wake-up times on school days appear to disrupt children’s natural sleep cycles. Maternal employment affects children’s sleep time in the summer, because children wake up earlier on days that their mothers work. But during the school year, maternal employment effects are dominated by school effects.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
This is the National Sleep Foundation recommendation for children ages 5–12 (http://www.sleepfoundation.org/article/sleep-topics/children-and-sleep).
See Carrell et al. (2011) for a nice summary of this literature.
As a result of this body of research, some school districts have pushed back high school start times.
The evidence on the effect of maternal employment on child outcomes is somewhat mixed, but most studies find negative effects. Baum (2003) and Ruhm (2004) find negative effects on cognitive development, while James-Burdumy (2005) finds positive effects. Other studies have found positive associations between maternal employment and the incidence of injuries and infectious disease (Gordon et al. 2007), and obesity (Anderson et al. 2003).
Other papers, by Cawley and Liu (2007b) and Fertig et al. (2009) have examined the mechanisms by which maternal employment might lead to child obesity. They focus mainly on the reduction in time spent preparing meals at home, which implies that children are eating more prepared foods (prepackaged or restaurant food). Bonke and Greve (2012) found no relationship between parental work hours and the amount of time children spent in health-related activities.
These questions were added to the survey, because cognitive testing revealed that respondents were not consistent in reporting secondary childcare at times when their children were asleep (some parents reported this time, while others did not). To ensure consistency across respondents, the ATUS asks the wake and sleep times and excludes any secondary childcare that was reported when all children under 13 were asleep. The ATUS does not ask about naps, which means that reported secondary childcare time includes times when children are napping.
The percentages of 6–8, 9–10, and 11–12 year-olds with older siblings are 36, 50, and 60, respectively.
There were only 172 children with married parents and a non-employed father, and 212 children living with single fathers.
I did not include father’s race and child’s race because they are highly correlated with mother’s race.
Stewart (2010) presents evidence that part-time employed mothers spend more time with their children and that they arrange their schedules to spend time with their children at times of day that enhance parent–child interactions. In contrast, full-time employed mothers appear to have very little flexibility regarding their schedule.
Unfortunately, there is no way to know if children attended summer school or some other activity.
To simplify both reporting and hypothesis testing I estimated two versions of each regression. The first regression includes the variables of interest (employment status, part-time status, etc., along with the interactions of these variables with the single mother indicator) plus each of these variables interacted with an indicator variable for summer. I reported the main (non-interacted) effects in the school year column. The second regression was estimated the same way except that the variables of interest were interacted with the school year indicator instead of the summer indicator. Analogously, the main (non-interacted) effects are reported in the summer column. The equations are not fully interacted because the coefficients on the control variables are constrained to be the same for both seasons.
Employed mothers in my sample work an average of 4.9 days per week.
As with the previous regressions, this regression was estimated over both school year and summer observations. For obvious reasons, only the school year coefficients are shown.
It is not clear whether it is more appropriate to enter an indicator for workday or workday eve (the day before a workday). The workday variable is included because it is not possible to identify days before workdays.
The implied 34 min effect on total sleep time is not inconsistent with the 57 min effect found in Table 3. The difference is that the school day variable in Table 3 does not account for the fact that children go to bed earlier on Sundays and stay up later on Fridays. When I reran the bed-time equation with the school day variable in place of the school night variable, the difference is 57 min (as in Table 3).
The National Institutes of Health (2011) recommends that school-age children and adolescents get at least 10 h of sleep per night. The National Sleep Foundation recommends that children ages 5–12 get 10–11 h of sleep per night (http://www.sleepfoundation.org/article/sleep-topics/children-and-sleep).
It is worth pointing out that using summer sleep patterns as a comparison group is potentially problematic. Children may be more physically active in the summer and may tend to sleep more. Working in the other direction, the daylight hours are longer in the summer, which may lead to children to sleep less.
For each sub-sample in Table 5, I computed the predicted values using the coefficients on the relevant indicator variables for mother’s employment status and whether the diary day was a school day or a workday. To obtain an average over the entire sample, I weighted the coefficients on the single-mother × employment interactions by multiplying these coefficients by the fraction of single mothers in the sample. The effect of the remaining control variables are evaluated at their sample means. I computed standard errors by first generating the standard error of the predicted value for each observation. I then averaged these standard errors over the relevant subgroup [for example, the children of full-time employed mothers on school days (nights)].
See Frazis and Stewart (2011) for an analysis of what can and cannot be estimated using time-diary data.
References
Anderson, P. M., Butcher, K. F., & Levine, P. B. (2003). Maternal employment and overweight children. Journal of Health Economics, 22, 477–504.
Banks, S., & Dinges, D. F. (2007). Behavioral and physiological consequences of sleep restriction. Journal of Clinical Sleep Medicine, 3(5), 519–528.
Baum II, C. (2003). Does early maternal employment harm child development? An analysis of the potential benefits of leave taking. Journal of Labor Economics, 21(2), 409–448.
Bonke, J., & Greve, J. (2012). Children’s health-related life-styles: How parental child care affects them. Review of Economics of the Household, 10, 557–572.
Carrell, S. E., Maghakian, T., & West, J. E. (2011). A’s from Zzzz’s: The causal effect of school start time on the academic achievement of adolescents. American Economic Journal: Economic Policy 3(3), 62–81.
Carskadon, M., Viera, C., & Acebo, C. (1993). Association between puberty and delayed sleep preference. Sleep, 16(3), 258–262.
Cawley, J., & Liu, F. (2007a). Mechanisms for the association between maternal employment and child cognitive development. NBER Working Paper 13609.
Cawley, J., & Liu, F. (2007b). Maternal employment and child obesity: A search for mechanisms in time use data. NBER Working Paper 13600.
Crowley, S., Acebo, C., & Carskadon, M. (2007). Sleep, circadian rhythms, and delayed phase in adolescents. Sleep Medicine, 8, 602–612.
Epstein, R., Chillag, N., & Lavie, P. (1998). Starting times of school: Effects on daytime functioning of fifth-grade children in Israel. Sleep, 21(3), 250–256.
Fallone, G., Acebo, C., Seifer, R., & Carskadon, M. (2005). Experimental restriction of sleep opportunity in children: Effects on teacher ratings. Sleep, 28(12), 1561–1567.
Fertig, A., Glomm, G., & Tchernis, R. (2009). The connection between maternal employment and childhood obesity: Inspecting the mechanisms. Review of Economics of the Household, 7, 227–255.
Frazis, H., & Stewart, J. (2007). Where does the time go? Concepts and measurement in the American time use survey. In E. Berndt & C. Hulten (Eds.), Hard to measure goods and services: Essays in memory of Zvi Griliches. NBER Studies in Income and Wealth. University of Chicago Press.
Frazis, H., & Stewart, J. (2011). How to think about time-use data: What inferences can we make about long- and short-run time use from time diaries? Annales d’Economie et Statistique (Annals of Economics and Statistics), 105/106, 231–246.
Gangwisch, J. E., Malaspina, D., Boden-Albala, B., & Heymsfield, S. B. (2005). Inadequate sleep as a risk factor for obesity: Analyses of NHANES I. Sleep, 28(10), 1289–1296.
Gordon, R. A., Kaestner, R., & Korenman, S. (2007). The effects of maternal employment on child injuries and infectious disease. Demography, 44(2), 307–333.
Hale, L., Berger, L. M., LeBourgeois, M. K., & Brooks-Gunn, J. (2009). Social and demographic predictors of preschoolers’ bedtime routines. Journal of Developmental and Behavioral Pediatrics, 30, 394–402.
Hamermesh, D. S., Frazis, H., & Stewart, J. (2005). Data watch: The American time use survey. Journal of Economic Perspectives, 19(1), 221–232.
Hofferth, S., & Sandberg, J. F. (2001). How American children spend their time. Journal of Marriage and Family, 63(2), 295–308.
James-Burdumy, S. (2005). The effect of maternal labor force participation on child development. Journal of Labor Economics, 23(1), 177–211.
Ruhm, C. (2004). Parental employment and child cognitive development. Journal of Human Resources, 39(1), 155–192.
Stewart, J. (2010). The timing of maternal work and time with children. Industrial and Labor Relations Review, 64(1), 181–200.
Acknowledgments
I thank Harley Frazis, Sabrina Pabilonia, three anonymous referees, and special issue editor Almudena Sevilla for their helpful comments. Any views expressed in this paper do not necessarily reflect those of the Bureau of Labor Statistics.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Stewart, J. Early to bed and earlier to rise: school, maternal employment, and children’s sleep. Rev Econ Household 12, 29–50 (2014). https://doi.org/10.1007/s11150-013-9182-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11150-013-9182-0