Methodology

---

Materials

We constructed a blocked design: one block was the control group and another was the treatment group. After the treatment group received six weeks of formalized spreadsheet training, both groups were administered post-tests. The post-tests have been validated by ETS and are considered an equivalent test to the pre-test. The ETS tests indicate no significant learning effect between the pre- and post-test. The post-test followed the same basic steps as the pre-test, except the participants were not required to complete the demographics questionnaire.

Participants and Training Process

The participants in the experiment were undergraduate students enrolled in one of two courses conducted at a large land-grant university: Survey of Accounting (control group), or Personal Computers in Business (treatment group). A total of 42 subjects participated in both the pre- and post-test with 50% (n=21) in each of the treatment and control groups. We used a blocked controlled experiment (Cook & Campbell, 1979) where the students enrolled in the personal computer class received the treatment (i.e., spreadsheet training). The sample was composed of 48% males and 74% sophomores. Participants had a mean Grade Point Average of 2.76 (based on 4.00 scale) with a standard deviation of .45. Forty-eight percent had no work experience involving computer usage. Ninety percent were full-time students (taking 12 or more credit hours/semester) and 70% were not working during the semester of the experiment. Students were free to discontinue participation in the experiment at any time. Seven participants failed to return for the post-test, and were removed from the sample, resulting in the final total of 42 participants.

At the end of the training period, students were required to take a spreadsheet content examination using a personal computer in which they had to create, change, or comment on several spreadsheet models. This synchronous, on-site testing environment was considered the most appropriate way to evaluate what the students had actually learned from the spreadsheet training. The participants in the treatment group averaged 86% on the spreadsheet exam and averaged 86% on the projects indicating that they had obtained at least a reasonable degree of proficiency in the required concepts and tools of proper spreadsheet design and implementation.

---