Characteristics in response rates for surveys administered to surgery residents.

Abstract

Surveys are important research tools that permit the accumulation of information from large samples that would otherwise be impractical to collect. Resident surveys have been used frequently to monitor the quality of postgraduate training. Low response rates threaten the utility of this research tool. The purpose of this study was to determine the standard response rate of surveys administered to surgery residents and identify characteristics associated with achieving greater response rates. A search of peer-reviewed literature published between September 2003 and June 2011 was performed with the use of PubMed with Medical Subject Headings: "internship and residency," "surgery," "data collection," and "questionnaires." For inclusion, articles must have described a survey given to active surgery residents within the United States. Surveys were evaluated based on the following criteria: population size, response rate, incentive use, follow-up use, survey format (online vs paper), and institution versus national. Of 433 initial results, 47 met inclusion criteria with a mean response rate of 65.3%. Surveys administered in paper format had a greater response rate compared with those given electronically (mean 78.6% vs 36.4%, respectively, P < .001). Greatest mean response rates were seen for institutional surveys compared with those given nationally (83.1% vs 42% respectively, P < .001). Our review demonstrated that paper surveys administered at the institutional level and during assemblies integrated into residents' schedules demonstrated enhanced response rates. The validity and generalizability of data collected through such surveys will improve as the aspects which dictate response rate are better understood and implemented.