This semester, FSU became the newest consortial member of Atlanta’s Census Research Data Center. Funded primarily by the College of Social Sciences and the Office of Research, the Florida State community can now use Census micro-data without paying lab fees, which can range upwards of $15,000 per project. There are currently 18 Census Research Data Centers in the United States, and outside of North Carolina’s Research Triangle the only one located in the southeastern United States is The Federal Reserve Bank of Atlanta.
So, what is a Census Research Data Center? The Center for Economic Studies defines Census Research Data Centers (RDCs) as U.S. Census Bureau facilities, staffed by a Census Bureau employee, which meet all physical and computer security requirements for access to restricted–use data. At RDCs, qualified researchers with approved projects receive restricted access to selected non–public Census Bureau data files.
To understand the true value of doing research with non-public data from the RDC, it’s important to note the difference between micro data and macro data, which is often referred to as aggregate data. When most of us use datasets for research or analysis, we’re looking at summary figures. For example, if you extract Census data for analysis, you’re typically looking at some sort of summary or aggregation for a specific geographic unit. These geographic units range from state, county, city as well as much smaller units such as census tracts and block groups. Regardless of unit of analysis, the data itself is a summarization of individual survey responses for participants in that specific area.
Aggregate data is quite useful for analyzing characteristics of a certain area as well as facilitating informed decision making. However, there are limitations to only having access to aggregate data. If we want to measure the impact that one variable, like educational attainment, has on another variable, like household income, we’ll need to look at the actual survey responses. This method of evaluating the link between two or more phenomena is referred to as regression analysis.
As you can imagine, there are several obvious reasons why Census micro data is restricted and not for public use. The most notable issue is the privacy of the participants. This is why Census RDC’s are housed in a secure location and have stringent standards regarding who can use the facility. In order for a project to be approved it must:
- Provide benefit to Census Bureau programs
- Demonstrate scientific merit
- Require non–public data
- Be feasible given the data
- Pose no risk of disclosure
When most people think Census data, they think of population statistics. While the Census is essential in providing population counts for the U.S., their data topics cover a wide range of areas like: economics, demography, and public health. The Census RDC also houses micro data on Longitudinal Employer-Household Dynamics, which is especially beneficial because it uses the same respondents and captures changes in behavior over a period of time.
My background is in the social sciences so I’m naturally excited about the possibility of analyzing any sort of survey micro data. However, access to the Atlanta Census RDC likely has significant utility for folks in the STEM fields as well as other disciplines. The ability to access and analyze micro data records from a survey that attempts to capture 100% of the American population provides researchers with endless opportunities for rigorous statistical analysis of our population.
If you’re interested in doing research at the Atlanta Census Research Data Center, please contact me at rjulian at fsu.edu and I’d be happy to talk to you about putting together a proposal. Now, go out there and get your hands dirty by taking advantage of this fantastic resource!