Hybrid Estimation: Using Probability Surveys with Social Media and Other Big Data (Cancelled)

This course has been cancelled.

This course is about developing statistical estimates that capitalize on the strengths of probability samples together with larger, often cheaper, data sources such as social media, administrative records, etc. Probability-based samples backed by theory and literature on best practices are becoming more difficult to implement with higher costs and lower response rates. Meanwhile, “Big Data” (e.g., administrative records, social media data) are more accessible than ever. A two-step hybrid design offers promise by allowing estimation from a combined set of probability and nonprobability samples. First, the weighted probability sample is used to calibrate the nonprobability sample to improve representativeness of the nonprobability sample. Second, the probability and nonprobability samples are blended to reduce sampling and nonsampling errors. Open questions remain on the best practices for both of these steps. In this course, we will briefly discuss the current issues researchers face with probability samples and Big Data; define hybrid or blended sample surveys in the context of a few examples; and, through review of the current literature, discuss the estimation challenges and solutions currently in use. The course will consider the practical implications of hybrid sampling on survey estimates including sampling precision, nonresponse bias, coverage bias, measurement error, processing error and specification errors in the integration process. We will use examples from the instructor’s research to help illustrate and demonstrate each of the methods discussed in the course.
 

This course will count as 4.0 CSS short course credit hours.

Instructor: Marcus Berzofsky

Marcus Berzofsky, DrPH, has been designing and analyzing surveys for over 15 years. He is currently a Senior Research Statistician at RTI International. Dr. Berzofsky has led large and small survey studies for multiple federal and state agencies. His research interests include designing more efficient sample designs and the impact those designs have in terms of total survey error. Additionally, Dr. Berzofsky is interested in methods to best sample rare populations and measure sensitive outcomes. He has published over 20 peer reviewed journal articles on these topics. Dr. Berzofsky received his DrPH in Biostatistics from the University of North Carolina’s School of Public Health in 2011.
 
 

Registration fees

• CSS Students: $30
• UNC-CH Students: $40
• UNC-CH Faculty/Staff/Postdoc/Others: $60

 
Cancellation/Refund Policy: A full refund will be given to those who cancel their registration no later than 10 days prior to the course. If you cancel within the 10 days prior to the class, no refund will be given. Please allow 30 days to receive your refund.

Zoom link for this course will be sent prior to the course. Registration must be made at least 3 days prior to the course date to receive the Zoom link.