Getting a Valid Survey Response Rate From 608 Academic Plastic Surgeons

PURPOSE: Web-based surveys save time and money. As electronic questionnaires have increased in popularity, telephone and mailed surveys have declined. With any survey, a response rate of 75% or greater is critical for the validity of any study¹. We wanted to determine which survey method achieved the highest response among academic plastic surgeons.

METHODS: All American Association of Plastic Surgeons (AAPS) members were surveyed regarding authorship issues. They were randomly assigned to receive the questionnaire through 1 of 4 methods: A) emailed with a link to an online survey; B) regular mail; C) regular mail + $1 bill and D) regular mail + $5 bill. Two weeks after the initial mailing, the number of responses was collected and non- responders were contacted to remind them to participate. The study was closed after 10 weeks. Survey costs were calculated based on the actual cost of sending the initial survey, including stationary, printing, postage (Groups B-D) and labor. Cost of reminders to non-responders was calculated at $5 per reminder, giving a total survey cost.

RESULTS: Of 662 surveys sent, 54 were returned due to incorrect address/email, retirement or death. 417 of the remaining 608 surveys were returned and analyzed. The response rate was lowest in the online group and highest in those mailed with a monetary incentive (Figure 1).

Initial cost of survey for each group (A-D) was: $0.25, $1.98, $2.98 and $6.98 respectively. However, given the differences in the response rates, and the cost of reminding non-responders, the final cost per response for groups A-D was $7.37, $9.16, $6.89 and $11.33 respectively (Figure 2).

CONCLUSIONS: We obtained statistically significant response rates (79% and 84%) using postal mail with monetary incentives and reminders. The inclusion of a $1 bill represented the greatest value and cost-effective survey method, based on cost per response. Despite the convenience and low initial cost of web-based surveys, this generated the lowest response. Given the greater number of reminders required, this was ultimately more expensive to implement than mailed surveys, even with the inclusion of a $1 bill. Although mailed questionnaires with the inclusion of a monetary incentive had the highest response rate, only with the addition of reminders did we achieve response rates above 75%.

Figure 1

Figure 2