By Michael Lopez, Ph.D., Co-Principal Investigator of the National Research Center on Hispanic Children & Families, and Todd Grindal, Ed.D., Associate at Abt Associates. You can reach them on Twitter at @Milopez1960 and @Grindato.
The Context of “Big Data”
You may be tired of hearing about it, but the big data revolution is upon us. Forecasts of big data’s impact range from enthusiastic praise about its capacity for global transformation to sober cautionary notes regarding potential misuse and unintended consequences. Regardless of which theory you subscribe to, there is no doubt that this exponential growth in the amount and type of recorded information has potentially profound implications for how we live our lives.
Nowhere is this truer than in public policy debates about the use of data collected by federal, state and local government programs. Government public assistance, social service, health, and education programs each collect a tremendous amount of information on children and families as part of their day-to-day operations. These include demographic characteristics of individuals who receive services; the timing, type, and intensity of services received; and indicators of individual well-being for families, parents, and children. On their own, the data collected by each separate public service agency can provide useful information on the reach and effectiveness of individual programs. When combined across data sources, these data can help provide more detailed portraits of family and community life that can inform policy and improve service delivery.
The Emergence of Integrated Data Systems
Over the last decade, government efforts at all levels have supported the development of Integrated Data Systems (IDS) that link administrative and program data, both over time and across different service agencies. Researchers have begun mining these integrated data and have revealed new insights regarding housing, education, and criminal justice programs. However, to date, researchers focused on Hispanic children and families have yet to fully take advantage of this new resource. This is important because, despite decades of research, substantial gaps remain in our understanding of the public service utilization patterns of Hispanic families with young children.
In the latest research brief from the National Research Center on Hispanic Children & Families (Center), we explore how the use of these IDS can help respond to the service needs of low-income Hispanic children and families. We provide a primer on the nuts and bolts of integrated data and a discussion of some of the specific questions that Hispanic-focused researchers might explore using IDS. For example, the enrollment of Hispanic children in center-based early care and education programs often is lower than that of other groups, although we still don’t fully understand why. These gaps in the knowledge base regarding utilization patterns include which individual, family, and/or community characteristics may facilitate and/or impede access, how utilization patterns vary by Hispanic subgroups, and how differences in utilization patterns may be associated with differential outcomes for children. Understanding these dynamics and differences in service utilization patterns is critical to designing and administering public service programs that are responsive to the needs of low-income Hispanic families and their children.
IDS have the potential to provide a comprehensive, timely, and cost-effective mechanism for examining these critical questions, particularly given that much of the data within have already been collected. Yet, in order for IDS to best support service delivery and programing for low-income Hispanic children and families, the information from the individual data systems must include relevant information that reflects the considerable diversity of the Hispanic population. The term Hispanic encompasses a tremendously diverse array of families that differ in terms of important characteristics such as country of origin, time in the United States, and English language proficiency, among others. As was noted in an earlier post on this blog, this information is not always collected as part of national surveys. The same is true of the administrative data collected by various government assistance programs. In order for IDS research to best support our understanding of low-income Hispanic children and families, it will be important for agencies to begin collecting more detailed data on the diversity of Hispanic families.
The mere accumulation of linked longitudinal information on children and families, on its own, is of limited use for the policy and practice communities. As Harvard professor Gary King has noted, the big data revolution is as much about analysis as it is about the collection of the data itself. As an example, Center researchers, in collaboration with researchers at Chapin Hall, are currently conducting analyses on the use of publicly-funded early care and education and related public assistance services among low-income Hispanic families with young children in Chicago. Bringing together information from nearly a half-dozen sources, this study will provide insights regarding patterns of public service use among low-income Hispanic families during the children’s early years. Equally important, this forthcoming IDS study will provide examples of how to approach the analysis of IDS data to address policy-relevant questions related to the service needs of low-income Hispanic children and families.
With increased coordination between researchers, policymakers, and program staff, the emergence of IDS may provide an important boost to the growing momentum of the big data movement and support programmatic and policy responsiveness to the needs of the growing population of Hispanic children and families.