Falsification

Interviewer falsification, the act by a survey interviewer of faking an interview or turning in falsified results as if they were the real thing, is a well-known, long-standing, and recurrent problem that has drawn occasional attention in the research literature since the early days of the field's development. It has traditionally been referred to as curbstoning, a term that captures the image of the interviewer, out on field assignment, who settles on the street curbing to fill interview forms with fabricated responses instead of knocking on doors to obtain real interviews.

In recent years, the problem has drawn renewed attention because the U.S. federal government's Office of Research Integrity (ORI) made clear in 2002 that it considers interviewer falsification in any study funded by the U.S. Public Health Service to be a form of scientific misconduct. Because that designation can invoke potentially grave consequences for researchers and their organizations, a summit conference of representatives of governmental, private, and academic survey research organizations was convened in Ann Arbor, Michigan, in April 2003 by Robert M. Groves (with ORI support) to compile “best practices” for the detection, prevention, and repair of interviewer falsification. This entry draws freely on the statement generated from those meetings, which has been endorsed by the American Association of Public Opinion Research and by the Survey Research Methods Section of the American Statistical Association. This entry defines falsification, discusses its prevalence and causes, and outlines methods of prevention and control. This entry also covers actions to be taken when falsification is detected, suggests that control methods should be covered in reports of survey methods, and considers which falsification events can be handled internally.