Increasing response to mailed questionnaires by including a pencil/pen

Emily White, Patricia A. Carney, Ann Shattuck Kolar

Research output: Contribution to journalArticlepeer-review

30 Scopus citations


Nonmonetary incentives lead to small increases in response rates to mailed questionnaires. However, inclusion of a pen or pencil, which may be a facilitating factor as well as a reward, has not been shown to improve response to health surveys in prior trials. In 2001 and 2002, the authors conducted two US trials in which a study-logo pen or pencil was randomly included in a second questionnaire mailed to nonresponders to a first mailing. In the first study, of 10,686 nonresponders to a cohort recruitment mailing, response to the second mailing was 55% with inclusion of a pen versus 40% without one (p < 0.001). In the second study, of 141 nonresponders to a pilot follow-up survey conducted 2 years after entry into a cohort, response was 43% with inclusion of a pencil versus 24% without one (p = 0.02). This 15-19 percentage point increase for mailing 2 translated to a 5-6 percentage point increase after the two mailings combined. In a simulated study of three mailings based on these studies, the overall response rate increased by 4 percentage points at no added cost through inclusion of a pencil in the second mailing. The additional cost of the pencil was compensated for by the reduced number of nonrespondents sent packets at the third mailing. This study supports including a study-logo pen or pencil in a second questionnaire mailing to nonrespondents as a cost-effective method of increasing response rates.

Original languageEnglish (US)
Pages (from-to)261-266
Number of pages6
JournalAmerican journal of epidemiology
Issue number3
StatePublished - Aug 2005
Externally publishedYes


  • Data collection
  • Epidemiologic methods
  • Questionnaires

ASJC Scopus subject areas

  • General Medicine


Dive into the research topics of 'Increasing response to mailed questionnaires by including a pencil/pen'. Together they form a unique fingerprint.

Cite this