Early Childhood Research Quarterly 21 (2006) 227–237
The effects of daycare: Persistent questions, elusive answers
Noam Shpancer ∗
Psychology Department, Otterbein College, Westerville, OH 43081, United States
Abstract
Despite nearly half a century of research on the effects of daycare on children, the fundamental question of whether non-parental
daycare adversely affects young children has not been answered conclusively. This article explores the conceptual and methodological
difficulties associated with daycare research and how the unique socio-historical place of the daycare issue in American society
interacts with inherent difficulties in the science–policy relationship to render a conclusive answer unfeasible.
© 2006 Elsevier Inc. All rights reserved.
Keywords: Daycare effects; Children; Science–policy relationship
Is non-parental daycare bad for children? Research on this question, as a well-defined area of scientific inquiry,
is approaching middle age, and it fits many of the attendant stereotypes. It is well established, attuned to its own
heritage, and given to occasional flurries of passion; but it is also sagging a bit, unsure about its sustained relevance,
and somewhat at sea about the actual meaning of it all. An interested, reasoned observer could have assumed that
this pressing basic question will have been settled, or else abandoned, with decades of intensive research (see Borge,
Rutter, Cote, & Tremblay, 2004); alas, like an aging rock band, the question of daycare effects insists on staging
periodic, dubious comebacks. The years, however, have not been kind to the field’s quest for empirically settling its
core question, as studies have continually yielded inconsistent, conflicting results.
For example, studies have linked daycare to children’s behavior problems (Belsky, 1999; Vandell, Burchinal,
Friedman, & Brownell, 2001), failed to find a link (Blau, 1999; Borge et al., 2004; Erel, Oberman, & Yirmiya, 2000),
or found that daycare is linked to a reduction in such problems (Denham & Burton, 1996; Field, 1991; Prodromidis,
Lamb, Sternberg, Hwang, & Broberg, 1995). Research has linked daycare participation to lower levels of aggressive
behavior in toddlers (Arsenio, 2004) and to higher levels of externalizing behavior in kindergarteners (NICHD ECCRN,
2003). The link between overall time in non-parental care and externalizing behavior disappears altogether by the third
grade (NICHD ECCRN, 2005). Bacharach and Baumeister (2003) concluded that, “there is no definitive answer to
the question and to whether there is a relationship between prekindergarten care arrangements and risk of subsequent
behavior disorders among children in kindergarten” (p. 529).
Regarding cognitive development, studies have found deleterious effects (Russell, 1999; Vandell & Corasaniti, 1990),
no significant links (Melhuish, Lloyd, Martin, & Mooney, 1990; NICHD ECCRN, 2000b; Votruba-Drzal, Levine Coley,
& Chase-Lansdale, 2004) and positive daycare effects (Andersson, 1996; Spieker, Nelson, Petras, Jolley, & Barnard,
2003). Vandell (2004) concluded that, “findings are mixed with respect to amount (and timing) of child care and
children’s cognitive, language, and academic performance” (p. 402).
∗ Tel.: +1 614 823 1515.
E-mail address: Nshpancer@otterbein.edu.
0885-2006/$ – see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.ecresq.2006.04.006
, 228 N. Shpancer / Early Childhood Research Quarterly 21 (2006) 227–237
Regarding socio-emotional development, research has shown that daycare hinders the quality of parent–child rela-
tions (Belsky & Rovine, 1988; Campbell, Cohn, & Meyers, 1995), does not hinder it (Booth, Clarke-Stewart, Vandell,
McCartney, & Owen, 2002; NICHD ECCRN, 1997), that the adverse effects are small and transitory (Harvey, 1999;
Scarr, 1997), or intermittent (NICHD, 1998). Early daycare has been linked to problems in parenting (NICHD, 1999a)
and to improvements in parenting interactions (Edwards, Logue, Loehr, & Roth, 1987; NICHD, 1997). Shonkoff and
Phillips (2000) concluded that, “when child care is found to be associated with the mother–child relationship, the
link is as likely to be positive as it is to be negative” (p. 310). Reviewing the literature on the effects of non-parental
care (NPC), Hagekull and Bohlin (1995) concluded that, “whether day care has detrimental effects or not in terms
of insecure attachment, non-compliance and aggression has been extensively studied, and no general conclusion can
be drawn at present” (pp. 506–507). Echoing a similar sentiment, Pierrehumbert, Ramstein, Karmaniola, Miljkovitch,
and Halfon (2002) stated that,
“Studies conducted in the past 20 years in many industrialized countries concerning the potential effects of NPC
yielded relatively inconsistent results. Whereas several early studies suggested it was harmful, most of the recent
ones have drawn more balanced conclusions. Overall, NPC may have positive or negative effects, depending
on the quality of the day care setting and on the developmental area assessed. Also, different children can be
affected in different ways and the effects of NPC seem to interact with family variables. Still, more studies are
needed . . .” (p. 385).
However, even this reasonable conclusion in fact glosses over important, and problematic, issues. First, the flow
of studies showing some negative effects has not, as is implied, dwindled over time (see Belsky, 2001; Brandtjen &
Verny, 2001; Creps & Vernon-Feagans, 2000). Second, as I will discuss below, the actual (causal) effects of daycare
on development depend in large measure on unknown variables, and on complex unmapped interactions between
variables. Third, the above conclusion seems to be saying, in essence: ‘we do not yet know how daycare affects
development; we thus require more study.’ This is problematic on two levels. First, for interested, thoughtful research
consumers—policymakers and parents among them—this state of affairs provides little guidance regarding care related
decisions. Second, this statement implies that we are on our way to finding a conclusive answer. But there are several
reasons to believe that the failure of daycare research to settle this basic question in fact constitutes the only viable
conclusion that can be derived empirically in this area, and understanding those reasons may aid interested parties in
both coping and decision-making.
1. Social science and the problem of truth
As social scientists studying a social institution, daycare researchers face inherent challenges. One does not have to
be a rabid postpositivist to acknowledge that the meaning and implications of a social institution cannot be separated
from the social context within which it is embedded. Social truths are ‘events in context’ (see Wachs & Shpancer,
1998), and context, of course, changes.
Social change, which is often rapid and unpredictable, works to continuously shift the terrain that daycare researchers
study. Findings and conclusions must be carefully qualified, as their validity is inherently contingent on multiple,
dynamic, socio-cultural parameters. Results that were valid five, 10 or 20 years ago may no longer be valid in the
present. As Pierrehumbert et al. (2002) noted, the fact that some recent studies (i.e., NICHD ECCRN, 1997) failed
to find the same adverse impact of non-parental care on attachment documented in earlier literature (i.e., Belsky &
Rovine, 1988) might be due to the increasing quality of child care and/or a reduction in parental guilt about daycare.
The above truism holds, of course, for other areas of research pertaining to the meaning and implications of social
institutions. For example, research has suggested that the adverse effects of divorce on children have been diminishing,
as divorce has become more common and accepted in the US (see Amato & Keith, 1991).
In the time it takes to solidify knowledge of a complex social phenomenon such as daycare, the nature (and context,
and meaning) of that phenomenon is likely to have shifted. In this way, despite its favored discovery ethos, scientific
inquiry often plays a frustrating and inherently losing game of ‘catch up’ with society, like those fabled golden gate
bridge painters who, once finished, have to immediately begin to paint the bridge all over again.
In addition, social science is not merely discovering and commenting on social reality, it is also a part of it. As an
institution becomes more visible and prominent in the societal consciousness, the implications of belonging to that
institution change, in part independently of the actual content of the coverage (see Pierrehumbert et al., 2002). Thus,
