The Future of Children, Princeton - Brookings: Providing research and analysis to promote effective policies and programs for children.

Accumulated evidence suggests that the large national surveys described above may yield underestimates of the true prevalence of self-care, for several reasons. Parents may underreport their use of self-care in part because of guilt and fear of legal consequences and in part because of problems with recall.^24,25 In addition, some of the large surveys do not collect data on occasional episodes of self-care.

Analyses of at least two national surveys mention relatively high rates of nonresponses to questions regarding self-care as suggestive of parents' reluctance to report fully on their use of self-care.^26,27 Further evidence of the impact of parental fear on reporting is found in a smaller survey of 447 rural, urban, and suburban parents.²⁸ This study reported that when parents were initially asked about their use of self-care, virtually all had replied that they did not leave their children alone. After a detailed explanation of the purpose of the research—to assess the children's skills in coping with simulated risks associated with self-care—the parents became more willing to disclose their use of self-care. Although this study does not provide direct evidence of the extent of underreporting due to fear, it does provide evidence that parents' initial reaction to questions about self-care is guarded.

Another study attempted to address the issue of underreporting due to problems with parental recall by conducting repeated evening telephone interviews with third-, fourth-, and fifth-grade children, during which the children reported on their care arrangements and activities in 15-minute blocks covering roughly the three-hour period after school. This study reported that 26% of third graders and 54% of fifth graders reported being alone at some time during the sampled time period.²⁹

None of the large national surveys measure occasional episodes of self-care. The 1990 NCCS counts only arrangements used at least once a week, and the SIPP reports only on arrangements in a "typical week."³⁰ Researchers recognize that many children's after-school hours are complex patchworks of different care arrangements, which may shift daily or weekly. A child may experience center-based care, organized sports or lessons, a baby-sitter, unsupervised time with peers, and self-care all in a single week. The 1990 NCCS represented significant progress toward capturing some of the complexity of care patterns, but while estimates from that survey of the number of children in self-care are the most comprehensive among the large surveys, they still do not capture occasional episodes of self-care—those that occur less frequently than once per week. Some of the smaller studies suggest that occasional self-care may be considerably more prevalent than regular self-care. One study found that while 8% of third graders experience self-care "regularly," 49% were left alone "occasionally."³¹

Finally, as an alternative to large national surveys that ask parents directly whether their children care for themselves, some researchers have attempted to estimate the incidence of self-care based on mothers' labor force participation rates. These estimates, which tend to be considerably higher than those based on survey data, are generally considered flawed. The Children's Defense Fund, for example, estimated that seven million children (21% of all children ages 5 to 14 years) were in self-care in 1982, based on the assumption that children in families with two full-time employed parents who are reported to be in parental care must actually be in self-care.³² This assumption ignores the evidence that in many families with two full-time working parents, one of the parents works other than a day shift. Many parents thus stagger their work schedules to keep their children in parental care.³²