An Experimental Test of the Effects of Survey Sponsorship on Internet and Mail Survey Response

Survey researchers have typically assumed that university sponsorship consistently increases response rates and reduces nonresponse error across different populations, but they have not tested the effects of utilizing different university sponsors to collect data from the same population. In addition, scholars have not examined how these effects differ for mixed-mode (web and mail) or mail-only data collection. To explore these questions, we conducted an experiment in spring 2012 with an address-based sample of residents from two states (Washington and Nebraska), using two university sponsors (Washington State University and the University of Nebraska-Lincoln) and two modes (a sequential "web-push" design versus a mail-only design). We found that within-state-sponsored surveys tended to obtain higher response rates than out-of-state-sponsored surveys for both "web-push" and mail-only designs. Our study also investigates the impacts of mode and sponsor on the representativeness of survey estimates.