Source bias is a statistical bias caused by the source of data being used to compute a statistical estimate. Source bias can affect simple descriptive statistics (e.g., mean, median, variance) as well as measures of association among variables (e.g., correlation coefficients, regression coefficients). The broader term method bias is often used instead of, or in combination with, the term source bias. A source of data has the potential to cause bias if it produces measurement error. Source bias is frequently discussed in the context of survey data since surveys often yield data with substantial measurement error, but scholars disagree about the extent to which source bias presents a major problem for social science researchers using surveys. This entry briefly describes two types of source bias—over- ...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles