I currently work as a Research Associate at the Eagleton Center for Public Interest Polling (ECPIP) at Rutgers University. At ECPIP, I help design our survey questionnaires, sampling strategy, and data sharing and visualization strategy. 

Data Visualization in Shiny Apps

Over the past year, I have revised our data sharing and visualization strategy to include Shiny Apps in R allowing quick visualization of survey results for clients. See examples below:

Improving Hispanic Outreach in New Brunswick, New Jersey 

My most recent project at ECPIP was to help revise our multi-mode survey of New Brunswick, New Jersey to ensure contact with hard-to-reach residents--particularly Hispanic residents. In previous iterations of the survey, we were unable to hit our demographic targets for New Brunswick's high Hispanic population, requiring weighting adjustments that increased our margin of error.  Working with our community partners in New Brunswick, I helped adjust our sampling design and outreach strategy after mapping Census data such as percent broadband access by Census tract and percent of households that speak Spanish as their primary language.  This enabled us to redesign our sampling approach to include a multi-method random sampling design--combining a door-to-door campaign, mail, phone, and text-to-web.

Our outreach to Hispanic citizens in New Brunswick improved by 12 percentage points compared to the previous iteration of the survey and Census parameters, which decreased our design effect and margin of error and allowed for more precise estimates. 

Percent Broadband Access by Census Tract in New Brunswick New Jersey (from US Census)