A virtual classroom experiment for teaching engineering economy

Engineering Economist, Winter, 2004 by William G. Sullivan, Janis P. Terpenny, Harpreet Singh

A two-tailed t test was used to test for a difference between the actual final scores of students in the control section and experimental section. At alpha = 0.05, a significant difference was observed between the means of actual final scores (t = 3.07, tabled [t.sup.*] = 1.66). Thus, students in the experimental group did perform better than students in the control group when the GE project score was included in the analysis.

Hypothesis 3: Are the Virginia Tech GPAs Different Between the Two Groups?

[H.sub.0]: Virginia Tech GPAs for students in the experimental group are not significantly different from students in the control group.

The null hypothesis suggests that the mean Virginia Tech GPA for students in the experimental section was not different from the mean Virginia Tech GPA of students in the control group. Table 4 indicates the descriptive statistics for this hypothesis.

A bi-directional t test was used to test for a difference between the control section and experimental section. Equal variances were assumed when performing the test, since students were drawn from the same underlying population. At an alpha = 0.05 level, no significant difference was found between the experiment and control group GPAs (t = 1.48, tabled [t.sup.*] = 1.66). However, at alpha > 0.1 levels, a significant difference is observed.

Summary of Spring 2000 Results

Table 5 summarizes the results obtained in the hypotheses tested in the Spring semester.

In summary, it was observed that Treatment 1 (use of GE projects to supplement learning) resulted in higher mean scores for the experimental group compared to the control group.

Fall 2000 Results

Descriptive Statistics

Table 6 summarizes descriptive statistics for the weighted final scores of both study groups. Although the nature of comparisons remains basically the same as that for the Spring 2000 section, some additional comparisons arise because students from both the experimental (10:10 a.m.) and control (9:05 a.m.) groups now have a project option. The difference between the two groups lies in the extent of use of Web-based materials. Hypotheses one through four presented below give the statistical analysis for the Fall 2000 experiment.

Hypothesis 1: Comparison of Mean Actual Score (No Project) for Control (9:05 a.m.) versus Experimental (10:10 a.m.)

[H.sub.0]: The mean actual score of the no project experimental group with Web-based supplemental materials was not significantly different from the mean actual score of the no-project control group who did not have Web-based supplemental materials.

The first comparison involved the control group section and the experimental group section to determine the impact of the Web-based supplemental material available to students in the experimental section. The descriptive statistics for these two groups are given in Table 7. The two-tailed t test gives a t value of 0.23, for tabled [t.sup.*] = 1.66. Thus, the hypothesis is accepted, and there is no statistical difference between the mean scores for the two groups of students.

Hypothesis 2: Comparison of Virginia Tech GPAs of Students in the Project (Experimental and Control Combined) and Non-Project (Control and Experimental Combined) Groups

[H.sub.0]: No significant difference existed between the Virginia Tech GPAs of students in the project and non-project groups.

Table 8 shows the descriptive statistics for this test. A two-tailed t test gives a t value of 3.96, for tabled [t.sup.*] = 1.66 at alpha = 0.05 level. Thus the hypothesis is rejected, and we conclude that there is a statistically significant difference between the students in the project and non-project groups. This finding acts as a guideline for the test to measure the effect of the project on student learning.