“…First, using different surveys across multiple institutions would confound survey and institutional variation, as survey salience can affect response rates (Groves et al, 2000;Heberlein and Baumgartner, 1978;Kojetin et al, 1993;Van Kenhove et al, 2002). Second, a similar argument can be made when comparing response rates across different methods of survey administration; response rates vary by the number of contacts (Dillman, 2000;Fox, Crask and Kim, 1988;Yammarino, Skinner and Childers, 1991;Yu and Cooper, 1983) and mode such as paper versus web (Cobanoglu, Warde and Moreo, 2001;Shannon and Bradshaw, 2002). Third, having similar samples and survey timing across schools is crucial, because response rates will vary as the sample and timing during the semester varies.…”