The 131st Annual Meeting (November 15-19, 2003) of APHA

The 131st Annual Meeting (November 15-19, 2003) of APHA

4174.0: Tuesday, November 18, 2003 - Board 3

Abstract #63139

Getting a foot-in-the-door and keeping it there: The sequential request strategy as a method to increase health survey participation over time

Michael Voloudakis, MA, MPH, Public Policy Research Institute, Texas A&M University, HC Dulie Bell Building, Suite 329, College Station, TX 77845, 979-845-6759, michaelv@ppri.tamu.edu and Robert J. Reynolds, MPH, Department of Psychiatry, University of California San Francisco, 2186 Geary Blvd., #103-104, San Francisco, CA 94115.

Survey data collection is a staple to public health research. However, response can vary greatly across topics and populations. Subject retention is a major concern due to the longitudinal nature of much health survey research. With participation rates historically declining due to many factors, it is important to consider strategies to maximize response. Even minor increases in participation have the potential for significant impact given the size, scope, and importance of such research. This project uses a sequential request strategy to increase participation in health related questionnaires over time. The study was a modified, longitudinal, foot-in-the-door paradigm. Subjects were asked to complete either a 1-page questionnaire (treatment) or a 6-page questionnaire (control) at Time 1. Subsequently, all participants were asked to complete one 6-page questionnaire each week for three consecutive weeks. Results showed a treatment effect between the groups at Time 1 (chi-square (1) = 13.0, p<0.01; OR=3.66, 95% CI = 1.55-8.67). At Time 2, results approached significance (chi-square (1) = 2.5, p=0.08; OR=1.77, 95% CI = 0.86–3.61). At Time 3 (chi-square (1) = 5.2, p=0.02; OR=2.31, 95% CI = 1.12 – 4.78) and Time 4 (chi-square (1) = 11.1, p<0.01; OR=2.78, 95% CI = 1.50 – 5.14) results were significant. An ANOVA model examining overall participation by treatment was also significant (df = 1, F = 24.4, p<0.001). Continuing analyses will examine the impact of demographic and personality variables, however even these preliminary results support the use of sequential request techniques as a method to increase longitudinal survey participation.

Learning Objectives:

Keywords: Survey, Public Health Research

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

Survey, Epidemiologic, and Clinical Methods

The 131st Annual Meeting (November 15-19, 2003) of APHA