Skip to main content icon/video/no-internet

A response rate is a mathematical formula that is calculated by survey researchers and is used as a tool to understand the degree of success in obtaining completed interviews from a sample. In probability samples, where the intent of a survey is to project the results of the data onto a population (e.g. all adults in the United States), statistical theory rests on an assumption that data are collected from every unit, or person, selected. In practice, it is extremely rare for any survey to achieve this perfect level of cooperation from respondents. In turn, survey researchers may consider, examine, and when necessary, compensate for potential problems that this deficiency presents.

Response rates, sometime termed outcome rates, have traditionally been a topic of great interest because they describe the amount of nonresponse in a given survey. In doing so, they provide an indicator that can be used to better understand threats to the validity of survey data. Response rates inform researchers of the proportion of their sample that did not respond and also may lend insight into the reasons selected persons (or units) did not respond.

Background

Although nonresponse has been studied since the 1940s, serious efforts to standardize the measurement of nonresponse have arisen only within the last quarter of the 20th century. Furthermore, the common use of standardized response rate measurements has not yet been fully realized throughout the survey research profession.

Traditionally, there has been a great deal of overlap and inconsistency in both the definitions and formulas used to understand the concept of response rates. These discrepancies present a difficulty to the survey research profession because they often confuse consumers of survey information. Using consistent outcome rates is important because it allows the level of nonresponse to be compared more easily between different surveys. This provides researchers and clients or other end-users with a meaningful target when planning the design of research. Equally as important, standard outcome rates offer an important benchmark for understanding how well surveys performed.

For example, a lack of consistency prohibits the accurate comparison of nonresponse between two unique surveys, obscures agreement in target levels of nonresponse in research proposals, and hampers methodological research exploring nonresponse error.

In response to the historical differences among response rate calculations, the survey research profession has gradually worked toward a uniformly accepted set of formulas and definitions for nonresponse. These efforts are now spearheaded by the American Association for Public Opinion Research (AAPOR), which maintains a series of definitions, formulas, and dispositions that are continuously updated to reflect new technologies and changes in the survey research profession.

Cooperation Rate; Leverage-Saliency Theory; Nonresponse; Nonresponse Bias; Post-Stratification; Propensity Scores; Refusal Conversion; Response

AAPOR Response Rates

AAPOR first published a series of response rates in 1998 for random-digit dialing and in-person surveys due to the concerted efforts of Tom. W. Smith and his colleagues. AAPOR based their development of the rates on the earlier work of the CASRO (Council of American Survey Research Organizations), which had published a set of formulas in 1982. Prior to that time, there had been numerous inquiries into the development of standards but no successful efforts put forth, at the association level, toward rate development.

...

  • 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

Sage Recommends

We found other relevant content for you on other Sage platforms.

Loading