Improving the self reporting of tobacco use : results of a factorial experiment

(Disponible en anglais seulement.)

Self-reporting of tobacco use has often been considered a legitimate way to estimate tobacco use. When a surveyor announces the topic of their survey, this may influence a person's response. In reaction to this problem, a study was conducted to help determine which questions may lead a person to deny their tobacco use, and why.

http://tobaccocontrol.bmj.com/cgi/content/full/12/2/178

Comment: An interesting comparison between the self-reported tobacco use rates in two studies in California, and how question introduction, wording and placement can influence responses.