Limitations on Random Samples

Student projects are for learning the process of doing research, and for this purpose a small sample size is perfectly acceptable, as long as students understand that the criteria for publication of a study are more rigorous. Students understand that most class projects aren't intended for academic publication. If a student wants to follow up on an interesting preliminary result, that would be an appropriate topic for an independent study.

Most  students are using their own unvalidated survey instruments. It is perfectly acceptable to use a smaller-than-usual sample the first time you generate data with an untried instrument to get some preliminary results to guide your research or "tweak" the survey instrument. This is done all the time in "real research" and is called a pilot study.

As a result of these concerns, random samples of students (and in some cases, employee groups) through the Institutional Research Office will be limited to the following sizes.

100 for non-course polling (newspapers, student organizations, etc.)
200 for class projects below the thesis level
250 for administrative office surveys
500 for thesis-level projects with specific rationale for sample size

Office of Institutional Effectiveness
204 W. Washington St.
Washington and Lee University
Lexington, VA 24450-2116
540.458.4181
Office Hours: 8:30 am to 4:30 pm, Monday through Friday
Contact Us
© Washington and Lee University, 
This Page Last Updated: Thursday, September 14, 2006