The 3rd week of CEDO 540 has us in working in teams creating a survey. My study group met this week and we put together a survey to assess teachers attitudes towards technology in the classroom. Once we determined that we would use
SurveyGizmo we set about formulating the types of questions to ask that will yield relevant data for our analysis. The group decided it needs to know if the respondents had any experience with the types of learning technology.
Our list included individual and collaborative software/hardware used in our MEIT program and asked teachers to share their experience and attitudes towards each of the technologies. Our group made sure to list specific examples of technologies that encompassed spreadsheet, slideware, videoconferencing, virtual classroom, concept mapping software and others.
A big pitfall to avoid is
survey response bias. It is only natural that a survey author possesses an innate bias or attitude about the subject matter. One needs to keep that in mind and strive for simple language that does not lead the respondents. A great example of survey response bias could be found in years' past on
Comedy Central's 'The Colbert Report'. Stephen Colbert would ask guests, especially if they were liberal democrats,
"Is George Bush a great president or the greatest president ever?" It was funny how guests would fumble and sometimes plead for a third or not applicable option.
On Stephen Colbert's show the bias was obvious, overt and overwhelmingly funny. It is hard to miss. Still, one needs to avoid having a subtle bias that even the author(s) of a survey may not be aware that they are exhibiting. It might even be worth having 2 authors with opposing views craft the wording of a survey together. Once the data has been collected comes the time for the data analysis and interpretation. Stay tuned for my Week 4 post when my study group reveals our findings from the survey. Below is a teaser from our survey using
Likert 5-level item questions.
 |
| Sample Questions from our Study Group Survey |
I think the hardest part of the survey, besides designing meaningful questions, was deciding on how to reach a representative sample. I'm sure we did not do the best job of that, but it's really hard to find a substantially representative group in a week! P.S. I loved the video :)
ReplyDeleteGreat example on bias. It is hard to create a survey without bias and as Sarah said to choose a representative group of participants.
ReplyDeleteHi Brian! Always love checking out your blog. The Colbert report was hilarious! Yes, it stops you in your tracks if the answer for you isn't on the survey. Our group found out that we tended to ask questions that didn't need to be asked, but we didn't come across the topic of bias. I think there's some real merit to adding this to a list of things to think about while creating a survey!
ReplyDeleteThanks,
Mary
Brian,
ReplyDeleteI am glad you found that video from the Colbert Report. I am a big fan of it and it is very apt! I would agree with you that survey bias can be a troublesome aspect of designing and creating these surveys. Our group had the issue of making sure our questions were necessary. It is kind of neat to see the different parts of emphasis each group is using with these surveys.