American Journal of Epidemiology Advance Access originally published online on June 15, 2006
American Journal of Epidemiology 2006 164(3):296; doi:10.1093/aje/kwj244
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Letter to the Editor |
RE: "INCREASING RESPONSE TO MAILED QUESTIONNAIRES BY INCLUDING A PENCIL/PEN"
George Warren Brown School of Social Work, Washington University, St. Louis, MO 63130-4899
(e-mail: psacco{at}gwbmail.wustl.edu)
In reading "Increasing Response to Mailed Questionnaires by Including a Pencil/Pen" (1
), I was interested in the strong effects found by the authors. These results are valuable, offering an inexpensive means of improving response rates. Unfortunately, the reported findings are questionable in that they contradict previous research in this area, which found no significant effects from pencil incentives. According to the authors, the reasons for the differences were unclear. Upon closer review, it is apparent that their results may have been biased by the sample selection.
White et al. (1) are clear that their findings reflect improvements based on a second mailing, but they report that, in both samples, participants had either participated in previous research or agreed to be contacted for research purposes. The participants, in essence, had agreed to participate in research on two previous occasions. In the study 1 sample, women agreed to research through inclusion in a mammography registry. In the study 2 sample, all individuals had already completed an extensive questionnaire for the Vitamins and Lifestyle Study cohort. Selection bias was created in the sample by including only individuals who had agreed to participate in research before the first mailing.
Such selection bias can create a sample that is demographically unique compared with the general population. In the article, no information was provided about socioeconomic status or the racial and ethnic background of the respondents. Goldberg et al. (2) demonstrated the role of culture, socioeconomic status, and region on study participation. Without a representative sample, it is difficult to assert that including a pencil will improve overall rates of response in a more representative sample.
I am hopeful that the potentially meaningful findings of these authors will encourage other researchers to replicate these findings with other populations.
ACKNOWLEDGMENTS
Conflict of interest: none declared.
References
- White E, Carney PA, Kolar AS. Increasing response to mailed questionnaires by including a pencil/pen. Am J Epidemiol 2005;162:261–6.
[Abstract/Free Full Text] - Goldberg M, Chastang JF, Leclerc A, et al. Socioeconomic, demographic, occupational, and health factors associated with participation in a long-term epidemiologic survey: a prospective study of the French GAZEL cohort and its target population. Am J Epidemiol 2001;154:373–84.
[Abstract/Free Full Text]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  E. White and A. S. Kolar TWO AUTHORS REPLY Am. J. Epidemiol., August 1, 2006; 164(3): 296 - 296. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||