Logos on Envelopes Can Reduce Survey Returns: An Experiment in California
Royce Park
UCLA Center for Health Policy Research
Matt Jans
UCLA Center for Health Policy Research
John Rauch
Westat
David Grant
UCLA Center for Health Policy Research
Sherman Edwards
Westat
In this small experiment, we manipulated the delivery envelope and presence of a packet insert in the 2nd mailing of an address-based sample (ABS) screener form asking for basic household information and a telephone number for a subsequent telephone survey. The experimental envelope differed from the standard envelope only in the addition of the survey sponsor's logo (emphasizing public health). Addresses were randomized to receive either the standard (blank) envelope or the experimental envelope (with logo). The experiment was conducted on a random sample of addresses from the USPS Delivery Sequence File (DSF) in two California communities, and used the California Health Interview Survey (CHIS) telephone interview, which includes questions about health behaviors, health status, and insurance status. Results suggest that traditional advice about avoiding logos in mail survey design may be sound. We explore differences between the two communities, and between English and Spanish respondents. Implications for other mail survey designs, particularly those using ABS designs in hard-to-survey communities will be discussed.