A SAS program was used to generate a list of random numbers, which selected the 2000 students
who were included in the main study sample for this survey (Dillman, 2000).
The minimum acceptable sample size for this study was determined based on target
accuracy, as more relevant than target power, because its purpose was to accurately estimate the
strength of the relationship between the variables. In addition, sample sizes required to achieve
target accuracy are larger than sample sizes required to achieve target power. Therefore, sample
size based on target accuracy included more than enough people to achieve target power. For
research questions one and two, the minimum sample size necessary to accurately estimate the
strength of the relationship between the variables when conducting bivariate correlations is 297
(Algina, Degrees, 2005). For research questions three, four, and five, the minimum sample size
necessary to accurately estimate the strength of the relationship between the variables when
conducting multiple linear regression analyses is 231 (Algina, Sample, 2005). Thus, a sample
size of approximately 300 would be acceptable to conduct all of the necessary data analyses in an
effort to achieve target accuracy of p2+ .10 where p = .35. In order to increase the likelihood of
achieving an acceptable sample size, keeping in mind the response rate of approximately 25%
achieved for Phase Two of the pilot study, over-sampling indicated a sample size of
approximately 2000.
Data Collection
A pre-notification letter was sent via e-mail on February 16, 2007 to the main study sample
informing them that they would receive another e-mail within a couple of days requesting their
participation in a web-based survey (Appendix M). Two days later, a cover letter was sent via e-
mail to the main study sample explaining the purpose of the study, inviting them to participate,
and directing those who choose to participate to a link to the web-based survey (Appendix N).
One week later, another e-mail was sent to members of the main study sample who had yet to