One week later, remaining non-respondents received a second survey packet.

Later, all non-respondents received an e-mail with a hotlink to an Internet survey.

Internet survey. A random sample of 200 students was chosen for an Internet survey.

Each student received an e-mail invitation to participate, including a hotlink to the Internet survey and simple instructions.

Three days after the initial invitation, a reminder message was e-mailed to non- respondents.

One week later, a final reminder message was e-mailed to all non-respondents.

Page 2

Overall, Axiom received a total of 709 completed questionnaires, resulting in a response rate of 43%. When reviewing results, keep in mind that a sample of 709 interviews is subject to a sampling error of + 3.68 percentage points (at the 95% confidence level). In other words, if one were to conduct the survey 100 times, 95 out of 100 times the results would fall within 3.68 percentage points (plus or minus) of the true population-wide figures.

The following graph illustrates the increased effectiveness and efficiency of the Internet survey methodology compared to the mail survey methodology (keep in mind that Contact 4 for the mail survey was an Internet reminder).

The report includes an "Executive Summary," which summarizes the research findings, and "Research Results," which offer a descriptive analysis of students' responses to all research questions, including differences and similarities between subgroups where notable. A copy of the survey instrument with aggregate responses is included in Appendix A. "Data Tables," furnished to HUHS under a separate cover, include all of the survey questions cross-tabulated by

a key set of demographic and attitudinal variables.

Page 3