Hey Charlie! To make this study more useful, here're some suggestions for isolating variables -- (yes, I'm playing devil's advocate here, but I want to point out how many conflating variables there are to some of these studies...PLEASE don't feel like I'm picking on you. I believe that a lot can be gained through qualitative messy methodology, but not as many generalizations can be made). So: First, I'd say you'll need to establish baseline proficiencies in some way: how they're doing in this subject when they come to you. How are you going to measure their test results at the end of the semester? You'll probably need a similar pretest to see what they come to you doing and knowing since previous knowledge is one variable. Another key factor could be the population in the two groups: is there a reason people had to take the class at different times? Did one section fill up quicker? There may be people in one section, say at 10am, who would have preferred, knowing that they work better then, to be in the 2pm section. There may be a substantial number of certain kinds of majors who have their other classes at one of the times so that could skew your results. My 8am section is heavy with football players, for example; a 4pm section would have practically none, as they are discouraged from taking classes that will interfere in practice time. Do they all feel that they took the class at their optimal functioning time, or where do they rate it, say, on a Likert scale? This could skew a section. Next, what is their previous level of computing functioning? Do they see computers themselves as distractors or are they so familiar with computing tasks that they are part of the picture already? You'll want to measure students' anxiety level, experience / use or current performance functioning to see how they are capable of using the computers. I'll bet that Sarah would be far less distracted by checking her email during class than many people would; I'll bet that she has filters set on her email already and mailboxes set up and she knows about what she'll find when she checks email. Even if you're watching to see what students are doing in class on computers, you probably won't be able to gauge how distracting that is unless you also have some baseline about how they currently use computers in the first place. Again, how do they process information now? How do they study best and how do they learn best? These questions might include whether they listen to music or watch tv while they study for tests. But they could also include such far-reaching factors as memory retention of information presented visually vs. auditorily. There may also be "manipulatives" learners in the class who will learn better if they are tactically engaged with something -- maybe even the keyboard -- that would throw off your results if you hadn't tested for along the way. I teach writing, and I find that people who are "touch" learners will learn better when their hands are doing something -- doodling, taking notes, keyboarding etc. would all conflate your results for manipulative vs. non-manipulative learners. What subjects do they see themselves "good" at presently? Do they currently have anxieties about your class subject or "resistances" to being in the class that may encourage them to seek distraction in any form, even if it includes daydreaming? How do they measure their own interest in the class and subject at the outset? This might be something you assume to be equal between classes, but it may not be. Some people want to get their "bad" classes out of the way; others put them off. If you have a bunch in an early section who took it because they're no good at it, it could skew the section. You'll want to monitor how they use class time. If they are in the computer lab, monitoring whether they use the computers during class is only the first level. Are they playing games? surfing? searching? working on a paper for another class? IMing? How would they rate their own levels of distraction for the tasks that they are attempting to do? How are you going to decide statistical significance in the test differences at the end? Are you going to quantify the results as groups, or measure individuals' learning qualitatively? Whew! There're still lots of factors involved that could skew results: more people with children in one class than in another, for example; how much time they spend on homework for the class could be another. How tired or practiced you are in presenting your information in one class or the other. You might want to "fix" a few class days for the study -- days that you're particularly going to test knowledge that was gained on only that day, and design a self-reflective questionnaire that asks about some of these variables that the students would take after the test. You could see if you find correlations across data that way, too. At any rate, it would be useful to your own generalizations if you set out assumptions ahead of time. Yours in constructive method implementation -- Deanya (bracing for the onslaught of criticism) :-D. On Sunday, October 15, 2006, at 11:03 AM, Charlie Balch wrote:
Great points. I happen to have a perfect opportunity to do a little research here. I'm teaching two sections of the same class. In one section, I'm lecturing in a computer lab and in the other in a traditional classroom. I do not discourage my students in the lab environment from using the computers while I lecture. (Those that want to yell at me for this should start a new thread.) Assuming that my lectures are the major factor in student learning, tests are meaningful, and that use of computers is a distracter, I'm going to compare the test results at the end of this semester.
Charlie http://charlie.balch.org