Instructional module in Fourier spectral analysis, based on principles of "how people learn"

Journal of Engineering Education, Apr 2003 by Greenberg, Julie E, Smith, Natalie T, Newman, John H

Some very telling comments actually came in responding to this question on Lab 2, where four students made comparisons (either implicit or explicit) to Lab 1:

* "Basically I just followed the lab instruction in this exercise. It's not so challenging as the first lab exercise. But I really enjoy the feeling of solving real-world problems using my knowledge."

* ". . . I thought that a little more freedom in design of a coder decoder would be good."

* ". . .at times, it seems a bit excessive to 'hold our hands' so much."

* "I liked the overall organization of the lab, but I would have done something a little differently. Instead of giving templates for all the code, I would have given a shorter lab assignment, but also less help. I think this way would make students learn better the details of the vocoders and of the way each Matlab function works."

These responses were unexpected, as Lab 2 was unchanged from previous years and the instructor does not recall any such complaints in the past. Unfortunately, because student responses had not been saved the previous year, this is only anecdotal. Even so, it does appear that the HPL-informed module used in Lab 1 raised student expectations, and students were disappointed when those expectations were not met in Lab 2.

V. DISCUSSION

A. Student Understanding

Results presented in Tables 3 and 4 and Figures 1 and 2 clearly show that students in the treatment group (who learned the material with the instructional module) obtained a better understanding of some key concepts than students in the comparison group (who did not use the module). We speculate that this is in part due to the Web-based tutorial's systematic presentation of individual variables and concepts, which had a mitigating effect on students' tendency to be overwhelmed by the interaction of multiple variables. The benefit of the instructional module is more pronounced for concepts related to window length and shape than for concepts related to DFT length. Future work should consider how to modify the tutorial in order to help students gain a better understanding of the concepts related to DFT length.

This improved understanding of Fourier spectral analysis must also be interpreted with two caveats. First, we developed the module during the summer and fall of 2000, after collecting the comparison data but before the treatment data. Significant faculty development occurred as a result of the module development process, which included explicit specification of learning objectives and key concepts. This unanticipated faculty development may have confounded our assessment. It is possible that the course instructor did a better job of teaching to the key concepts for the treatment group. Second, in order to accommodate the HPL framework, we added a four-hour lab session, doubling the amount of lab time spent on the topic. So, it is possible that better teaching and more class time contributed to the improved understanding demonstrated by students using the module.