5
Approaches to Data Collection
In addition to discussing the issues and frameworks that should guide the substance of future research on child care for low-income families, participants at the third workshop addressed the process of research, asking what approaches to data collection are needed to generate useful answers to contemporary questions about child care.
Participants agreed on the value of encouraging research that employs different methodologies—short- and long-term studies; large and small datasets; analyses of administrative data; collections of qualitative data from providers, parents, and administrators, for example—because each approach fulfills different goals. The question is no longer which method to use, but how different methods can be used in conjunction to complement one another, they said.
The participants also stressed the need to be more deliberate about integrating different methods within single studies, as well as about coordinating analytic work and the dissemination of findings across studies. One participant noted, for example, that different analytic strategies applied to the same data can produce different conclusions. Multifaceted studies that integrate ethnographic and experimental methods, build questions into administrative databases, and embed intensive, focused studies into national surveys—as was done with the JOBS (Job Opportunities and Basic Skills Training) evaluation (Moore et al., in press) and the Teenage
Parent Welfare Demonstration (Aber et al., 1995)—were noted repeatedly as promising avenues to pursue.
Trade-offs were also highlighted between initiating new research projects and supplementing existing studies and between funding major experiments or longitudinal studies of child care and supporting lower-cost, smaller-scale projects. For example, noting the significant benefits of high-quality child care, one participant suggested a new longitudinal study that looks specifically at low-income children in widely varying qualities of child care, but others disagreed, noting the adequacy of existing studies—notably the NICHD Study of Early Child Care on child care quality (National Institute of Child Health and Human Development, 1995). Another participant suggested studying the differences between communities that vary in their supply of high-quality child care for such outcomes as poverty rates, welfare receipt, and school readiness. More modest suggestions included adding a few questions to ongoing national surveys (e.g., the Early Childhood Longitudinal Survey, the Survey of Income and Program Participation, and the Panel Study of Income Dynamics), supporting site-specific studies of local variation in child care markets, and identifying outputs of programmatic and policy initiatives (e.g., number of licensing code violations) that are hypothesized to affect outcomes for children and families. In this context, one participant noted the benefits that could be reaped if families' employment and child care histories, with dates, were added to existing surveys to track co-occurring patterns of employment and child care.
Given that state and local policies are likely to have a growing influence on the child care that is available to low-income families, the participants devoted considerable time to discussing the value of directing research toward state and local evaluation efforts, as well as to the importance of identifying research opportunities that are presented by state and local policies that affect child care. Examples cited at the workshop as models for the future include research projects that have studied the effects on the quality of care and child outcomes of naturally occurring changes in state child care regulations (e.g., the Florida Quality Improvement Study—Howes et al., 1995), and embedded studies of child care in statewide welfare reform experiments (e.g., the GAIN Family Life and Child Care Study —Meyers, 1992).
On a related topic, participants also discussed the value and shortcomings of state and local administrative databases as a source of data on child care. While common in the welfare reform literature, the child care field has neither mined nor supplemented administrative data as a poten
tially useful research strategy. Speakers noted opportunities to forge partnerships between academic researchers and federal and state child care agencies, resource and referral agencies, and others who manage local databases. They also cited the need for efforts aimed at improving the comparability of data across local and state databases. One participant suggested that a good starting point would be to select a couple of states with relatively advanced data systems to try out strategies for enhancing uniformity and cross-site ties. But because states have different systems and different administrative needs, one participant cautioned, researchers should be cognizant of the difficulties involved in this type of work even when good intentions prevail.
Other needs for improved integration and collaboration were also highlighted by the workshop participants. If efforts to understand what is happening in and because of child care are to be placed in the context of other influences on children's development and parents ' efforts to provide economic support for their families, then it is essential that research on child care become much more closely articulated with research in related fields—research on poverty, Head Start and early childhood and elementary education, child health, community and neighborhood influences, labor economics, and family structure. Encouraging interdisciplinary research is one means of bringing multiple vantage points to bear on child care research; major strides in this area—notably between psychologists and economists —have revealed both the opportunities and tensions that characterize efforts to bridge disciplines.
Participants agreed that a central challenge for the future is to identify effective and enduring mechanisms for fostering integrative research on child care—not only across disciplines, but between researchers and practitioners, between research that is funded by public agencies and private sources, and across states. They also agreed that child care research needs to be viewed and planned as a coherent enterprise, rather than as a collection of isolated research endeavors, and that it should be logically sequenced and designed to address the important issues that emerge over time.
Several participants questioned the meaning of integrative approaches to research in terms of funding mechanisms for child care research. They also raised questions about effective incentives for collaborative research, barriers to the cross-agency development and funding of research agendas, and ways to encourage continued dialogue rather than one-time collaborations on single studies. Participants suggested looking to other fields to learn lessons from attempts to establish more coherent research enterprises.