The Ask the Analyst series is a deep dive into the data by those most familiar with the CO APCD – the analysts themselves. We’ll hear about their experiences with recent analyses and answer any pressing questions that come up. Have a question for the CIVHC Analyst Team? Email it to info@civhc.org.
Name: Dagmar Velez, MS, Health Care Data Analyst
Client Name: National Bureau of Economic Research (NBER) and the State of Colorado
Project Name: Creating a Geocoded Socio-Economic Background Score Using American Community Survey Data
A geocoded socio-economic background (GSB) score was created to characterize members in the Colorado All Payer Claims Database (CO APCD). The GSB can then be used in various types of research.
- What were your first steps when beginning this analysis?
I worked with Maria De Jesus Diaz-Perez, Director of Research and Performance Measurement, to find relevant academic articles on the subject and identify the variables that would work best for our specific use case. Then we went through the available data on the American Community Survey (ACS) website to finalize our variables.
- Did you need to take on any specific considerations based on the data?
The academic article we used as a base for our calculations made it seem like all ACS variables used were continuous, but we found a couple of the income-related variables to be binary. We ended up adding the outputs of those as ‘0’ or ‘1’ to our composite score (before scaling), which worked out fine in the end!
- What challenges did you encounter while performing the analysis? How did you overcome them?
As with most survey data, the ACS database is difficult to maneuver for first-timers like myself. The site is not intuitive on where to go for a specific set of variables, but it’s very user friendly for high-level views. However, after much digging (and with Maria’s help) I can confidently say I can extract the data I want from the ACS website!
- Without delving into results, did anything surprise you about this analysis or the process of executing it?
I was surprised by how many members had no address or a low-quality address (eg. only a PO Box listed) in the CO APCD. This meant that we couldn’t create a GSB score for them, unfortunately. An important thing to consider when using this score since there may be more members with a low GSB score that have no address listed.
- What did you learn while performing this analysis?
I liked reading about some of the use cases of researchers using the ACS data! For example, the New Orleans fire department used ACS data (specifically age of structures, length of time the householder has lived in structure, and the household’s ratio of income to the poverty level) to estimate the likelihood that homes in an area were missing smoke alarms which helped them identify which areas to target installing them.