Widening the Lens: Using Federal Administrative Data as a Resource in Education Research

• In Event: Leveraging Federal Data in Education Research: Engaging With Administrative and Complex Sample Data

Abstract

Objective
The objectives of this presentation are (1) to broaden the understanding of the types of Federal data that are available and the regulations governing their use and (2) to inform researchers how to access these data and to link across datasets in ways that enhance context and increase explanatory power.
Framework
The Federal government collects data on many topics, from education to health. Although many education researchers are familiar with Federal statistical data, administrative data has been less accessible. Administrative data are data collected by government entities for program administration or regulatory purposes. Analysis of administrative data has led to improvements in grant processes, such as the Department of Education’s Investing in Innovation (i3) program. As part of the Open Data Initiative, the Office of Management and Budget (OMB) has encouraged Federal agencies to improve the accessibility and use of administrative data for statistical purposes (OMB, 2014b). Furthermore, the President’s FY 2016 budget provides specific guidance to make administrative data more accessible. This level of transparency and accessibility must also be balanced by privacy laws that govern disclosure risk and constraints in the current infrastructure of Federal information collection systems.
Method of Inquiry
This is a content analysis of policies related to the use of Federal statistical and administrative data. Quantitative data will be used to demonstrate the implementation of these policies.
Data
Data include both qualitative and quantitative data. Qualitative data include official policy documents, such as the President’s budgets (FY 2010 to FY 2016), laws, and statistical policy documents from OMB and Federal agencies (e.g., OMB, 2014a). Quantitative data include the Early Childhood Longitudinal Study – Kindergarten cohort (ECLS-K) public use data file and administrative data from the Federal Communications Commission’s Universal Service Program (e-rate).
Conclusions
Administrative data can be linked to currently available statistical data to evaluate educational programs. This can yield an excellent opportunity to assess funding streams directly tied to Federal programs and policies. Because most Federal information collections have intentional sampling strategies and large sample sizes, they provide users the ability to make population-level estimates and to detect effects for subgroups. By linking data through consistent identifiers and georeferenced data, researchers can also leverage contextual information, such as local poverty rates, which may influence outcomes. However, because of limitations in infrastructure and privacy laws, current access to some datasets, particularly administrative data, can prove challenging to even advanced data users.
Significance
The use of Federal data, both statistical and administrative, can serve to make research more policy relevant and contextually nuanced. By integrating these datasets into research programs, education researchers will gain power to answer questions – such as the critical “what works for whom?” – about program effectiveness.

Author

• Deanne W. Swan, National Center for Education Statistics