CNSTAT Releases Second Data Infrastructure Report in Series Intended to Modernize Data Collection

The National Academies of Sciences, Engineering, and Medicine (NASEM) Committee on National Statistics (CNSTAT) has released its second in an ongoing series of data infrastructure reports intended to modernize and enhance data collection (previous COSSA coverage). This latest report, Toward a 21st Century National Data Infrastructure: Enhancing Survey Programs by Using Multiple Data Sources, builds off the first by discussing how using multiple data sources beyond sample surveys could improve the data collection for statistics and social and economic research, as well as data equity.

The report acknowledges several data sources that could improve data collection, including probability samples, administrative records, and private-sector data. CNSTAT also recognizes the need for quality control of data collection and recommends establishing a new framework of quality standards and guidelines to ensure quality control. The report emphasizes that using data from multiple different sources will improve data equity through piecing multiple data sets together to create a well-rounded, complete data set. The report adds that there are four key ways that these recommendations will improve data collection:

1. Creating a framework for quality control of data sources is expected to target populations that have been underrepresented, either from under sampling or lack of response.
2. Using multiple data sources is expected to reveal a link between survey data and other data sources to create disaggregated statistics.
3. Through expanding resources outside of sample surveys, these recommendations are expected to produce statistics for small populations that are often missed due to survey sample sizes.
4. By using administrative data, agencies will be able to create data products and produce statistics from the data.

CNSTAT emphasizes the need to focus on data equity when modernizing data infrastructure to ensure that underrepresented populations are included in data and statistics. The final report is expected to focus on data privacy and avoiding harm when collecting data. Follow COSSA’s continued coverage on CNSTAT’s Data Infrastructure report series.