Saturday, October 3, 2009

Ethical Considerations of Surveys

Wednesday's facilitation about surveys helped me to further understand how much work surveys can be. They might take many, many revised revisions before the questions are as perfect as they can possibly be in order to avoid leading respondents to certain answers. It seems that one way researchers can avoid, as much as possible, creating biased surveys or questions that lead participants in a certain direction is by using a combination of questions. Maybe using both fixed and open-ended questions will allow for more accurate responses and interpretations of those responses by the researcher. Of course, even if a researcher uses mixed question types, the questions might still be biased or leading. However, one way to prevent unethical behavior with surveys is probably to have other researchers examine the questions or the interpretations the researcher has about the results. If it's even possible, like with interviews, the researcher could try to locate some of the participants after the survey is complete to see if those participants feel that the research has given accurate, unbiased interpretations of the responses. For example, if I surveyed 5 different composition courses, I could attend those courses to show the results I found. The students might have some kind of feedback as to whether or not they feel the interpretations accurately represent their answers. Maybe this would be impossible to do, but surely it is possible to have colleagues examine the research a person has done.

As for my own research, I'm sure I am biased in some way. I'm interested in understanding if teachers respond differently to students' written work in online classrooms compared to face-to-face classrooms. I'm interested in this study because I want people to understand that online instruction is valuable for many reasons--so I guess that is a bias right there. However, I've never taught an online course, so I actually really don't know what to expect to find if I do a study on this issue. I think that is helpful--because I don't feel that I would be going in looking for something specific. I have an idea of the different tools teachers use to respond to students in online courses, but I have no idea how those tools affect the kinds of response students receive about their written work.

I can see how for many people it might be tempting to reject certain responses from surveys that don't fit their preconceived ideas. I suppose we can all admit that it might even be tempting to do this, especially if we have invested a lot of time and energy into a project and feel that we really have something to prove (like an issue we care strongly about). However, I think it is very important to remember that not proving our point is also important. For example, maybe I find that students receive exactly the same kind of feedback in online courses as they do in face-to-face courses. Well, that says something very important about the teachers in our field and how students are learning. Maybe I find that teachers do respond differently to students in online courses. That would lead to more research about why this happens, what needs to change, or what face-to-face courses can learn from online courses. There's really a place to contribute to the field no matter which way my results lead me. But, in order to ensure I don't fall into the temptation to choose only the results that matter most to me, I can always ask colleagues to check my work, give their opinions, and challenge my results. Second and third and even forth opinions are very important. Plus, they might even lead to ideas that I never thought of before. Really, all work we do is collaborative, and knowledge is never "created" by just one person. It's important to have others review the work I have done for so many reasons.

1 comment:

  1. I love your idea to set up a feedback loop with survey data. I think too often we see survey data as something that belongs to researchers rather than something that could be further explained, questioned, or expanded by the participants involved with the survey. What a wonderful idea to use the survey data to feed a focus group, for example, or as a jumping off point for a round of interviews. Great idea!

    ReplyDelete