After listening to our last two guest speakers - and trying to incorporate what I have been taking in through our readings, I can't help but to want to immediately make better the pilot study I conducted last year. As I am preparing the survey instrument for my dissertation study, I am anxious to apply (already, even in my naivete) the proper types of questions that will help yield answers that will be truly meaningful. For example, my pilot study included rather closed, yes or no, or multiple choice answers. I asked primary care physicians if they used computers and other electonic devices to get information to give to their patients. I asked them questions which basically gave me numbers, yes/no answers which I ended up quantifying and graphing showing statistics in my final paper RATHER than a more qualified approach. NOW -- ah-ha moment -- I'm seeing how I need to shape my questions. I need to ask them to give me their own words as to how they approach "information gathering" for their patients. Studying information seeking behavior - and gathering the types of scenarios I am anticipating is much more than a"Dr. X does this and Dr. Y. does that" report. I want to see results like, "Dr. X does this BECAUSE . . . " and "Dr. Y. does that IN ORDER TO . . ." and come back with a true reflection of their behaviors. I hope this makes sense to someone else out there because, for me, I'm seeing my project taking on a much clearer, more defined route than I thought I was on. Thank goodness!
On another note, I'm going to be implementing a web survey and gathering results via e-mail, so I hope to be able to report on my progress and learn more about the process as this class continues.
~MJD
posted by MJ @ 4:42 PM
4 comments
links to this post
4 Comments:
Sometimes health care systems have their own pre-printed information to provide to patients. I'd include a question about "does your system provide pre-printed patient information?" i.e. brochures etc.
also, I know in the VA there is a computer for patient use in the library that is FULL of information. Oft times I contacted the librarian and sent patients to her directly. It was one way of opening a portal for the patient to a vast amount of information. Depends on the patient too, whether they can 'process' the information. I've found that even with my older patients MANY of them have computers at home. I'd ask if the physician/provider provided a website address for them to go to to additional information. A lot of times the providers don't have a printer in their office where they are seeing the patients to be able to print info out for a pt. Myself, I have sent pts to several websites I thought were good. Particularly with a new diagnosis. That's another thing...you may want to ask if the patients were newly diagnosed when the providers gave information.
I agree about going back and perhaps reformulating a research question after having done the readings for the class. It has to really snap...and pull all the pieces together...I thought I had / and have a good "topic" for dissertation...but my question needs to totally experience a real punch!!! That will help solidify it for me. The more we read and get those 'ah ha' moments, the more I feel I don't know anything at all!!
Mary Jo,
I think you've touched on something that is really basic to the discussion about qualitative vesus quantiative and it has to do with how you're shaping the questions.
I'm beginning to formulate a notion that the choice of research method (despite any outside influences like obeying the will of a dissertation committee)is dependent not only on the types of research questions we want to ask, but also on how much we want to direct or specify the types of answers we get.
It seems to me that a lot of quantitative methods try to determine ahead of time what possible answers there might be. For example, in survey research you spend a lot of time coming up with questions that have a discrete number of possible answers. Then you administer the survey and report the frequencies and percentages and maybe do some correlation work to see if the answer on one question is somehow related to an answer on another question. Statistical methods are wonderful for this type of analysis. Of course, theoretically, your predetermination of the possible answers is supposed to be based on the literature review you do. Prior research is supposed to preclude possibilities so that you don't have recreate the wheel or to point out possible answers that haven't had enough research, yet.
In qualitative research, I see the researcher still having an overall research question or questions in mind (that's absolutely essential), but, by selecting a qualitative method, choosing to not try to define all the possible answers ahead of time. By asking the qualitative-type questions that can generate much richer (although much more unstructured) results, the researcher can sift through that data to find possible answers that he or she never imagined.
I see the choice as determining where the researcher wants to put a lot effort -- before or after data collection. Should I put most of my effort up front and try to anticipate all the possible outcomes of the data collection so that my analysis and results are very straightforward or should I be open to the possibilities that bubble up as I do my iterative data collection and spend more time thoroughly analysing the data to discover the unanticipated answers? It really depends on what I want to discover.
Thank you for your thoughts and comments. I wish I would have taken a stronger methodology course about a year ago -- but, all is going to work out for the best now, I can see.
Post a Comment
Links to this post:
Create a Link
<< Home