Integrated Virtual Learning System for Programmable Logic Controller*

Journal of Engineering Education, Apr 2004 by Hsieh, Sheng-Jen (Tony), Hsieh, Patricia Yee

* Test 1 (before lecture) with Test 2 (after lecture);

* Test 2 (after lecture, but before module) with Test 3 (after module); and

* Test 1 (before lecture) with Test 3 (after module).

In cases where subjects were not present for all three tests, we only used complete pairs. So, for example, if a subject took Test 2 and Test 3, but was absent for Test 1, the subject's data was included only in the Test 2 vs. Test 3 analysis. The analyses revealed that for both the long and short module samples, the null hypothesis was rejected for all three pairs of tests. This suggests that the lecture caused a significant improvement in learning, and the timer instructions module subsequently caused further significant improvement in learning. Tests of means and variance were also performed to see if subjects in the long module and short module samples performed differently. Students in the longer module sample performed slightly better than those in the shorter module sample. However, there was no significant difference between the sample means and variance. Therefore, we can conclude that there was no module length effect on test scores. Tables 4 to 6 summarize the test statistics, critical value and conclusions for each test, where the null hypothesis is [mu]^sub d^ = 0.

5) Results from the Opinion Survey: We also computed means for the two groups on the opinion survey. Figure 5 summarizes these results. Student ratings were positive for all items, and there was no significant difference in ratings between the two groups. In general, students felt that the module was interactive, relevant, adequately paced, and easy to use and understand. We again performed tests of means and variance to see if there was significant difference in ratings between the long and short module samples, using the data from the question about "Ease of Use." In this case, students who took the shorter module rated the lesson's "Ease of Use" slightly more positively than students who took the longer module. However, the differences in means and variance were not statistically significant.

6) Student Comments: Student comments can be summarized as follows. Several students in the "long" group telt that the XAIDA-generated practice was repetitive and sometimes difficult to understand. Many students liked the timer animations; they felt being able to visualize how the instruction worked was very helpful. Some students commented that they would like to have examples of real-world timer applications (i.e., not just switches and lights). Overall, the students thought the module was helpful and supplemented the lecture well.

7) Discussion: Based on these results, we concluded first, that the Virtual PLC module on timer instructions was instructionally effective, and that students' subjective impressions of the system were positive overall. Second, there was no significant difference between the long and short module samples, which suggested that using short XAIDA lessons is a good strategy for saving time while still preserving the advantages of adaptive instruction. Finally, the most popular instructional activity appeared to be the animations, probably because they helped students to visualize timer functions. Thus, continued use of animations to illustrate command functions and other complex concepts appeared to be a good instructional strategy.