Biostatistics Journal Club: The Big Data Paradox and its Implications for Clinical Research

Surveys are an important tool for quantifying public behavior and opinions, but their generalizability depends fundamentally on the minimization of bias in the identification of their target populations. This talk will discuss how increasing sample size can magnify survey bias as conveyed in surveys of first-dose COVID-19 vaccine uptake in U.S. adults in 2021 from two large surveys, Delphi–Facebook and Census Household Pulse, compared to a benchmark from the Centers for Disease Control and Prevention. The study highlights that data quality matters more than data quantity, making the case that compensating the former with the latter is a mathematically provable losing proposition. Suzanne Dahlberg, PhD, of Boston Children’s Hospital will present.

Suggested Reading:

Unrepresentative Big Surveys Significantly Overestimated U.S. Vaccine Uptake

Speaker:

Suzanne Dahlberg, PhD