Choosing MeasuresPosted on January 20, 2009 Liliane Windsor (bio) discusses finding culturally appropriate measures. |
You must have Flash Player version 10.2 or newer installed. Please visit http://www.adobe.com/go/getflashplayer to download and install the latest version of Flash Player.
I develop a research project or a question that I'm interested in, and then the next step is to figure out what's going to be the methodology and so forth. Once you decide on that, you go to the literature to figure out what is available out there in terms of measuring the constructs that you need to understand for your research question.
And sometimes you'll find them, sometimes you don't find them. Sometimes you find them, but they are not that great. Sometimes you'll find them; they have been validated with a population, but not with your population. So you have to think about all of those things. And if you are lucky enough to find something that's already there, that's great. You can go ahead and use it, and if not, then there are a couple of ways you can go about that.
If you have something that has already been tested with a different population, you can look at it, perhaps you can modify some things to adapt it to use with your population. And then when you do your research, you can test the psychometrics of it. You can kind of incorporate that into the research and collect data for those purposes as well.
And then, the other thing you can do, if you don't find anything, you can try to do something from scratch. Now, sometimes, when you're developing a scale, you might find that it has been developed, not necessarily with input from the actual population. And I kind of like that, so usually, what I tend to do in developing my scales, I always go to the actual population. And I do some sort of qualitative research with them in terms of developing the items, making sure that everything is okay, and then I do the quantitative aspect of it in terms of testing the scale to see if it works.
And sometimes you'll find them, sometimes you don't find them. Sometimes you find them, but they are not that great. Sometimes you'll find them; they have been validated with a population, but not with your population. So you have to think about all of those things. And if you are lucky enough to find something that's already there, that's great. You can go ahead and use it, and if not, then there are a couple of ways you can go about that.
If you have something that has already been tested with a different population, you can look at it, perhaps you can modify some things to adapt it to use with your population. And then when you do your research, you can test the psychometrics of it. You can kind of incorporate that into the research and collect data for those purposes as well.
And then, the other thing you can do, if you don't find anything, you can try to do something from scratch. Now, sometimes, when you're developing a scale, you might find that it has been developed, not necessarily with input from the actual population. And I kind of like that, so usually, what I tend to do in developing my scales, I always go to the actual population. And I do some sort of qualitative research with them in terms of developing the items, making sure that everything is okay, and then I do the quantitative aspect of it in terms of testing the scale to see if it works.
Viewing Preferences
|
Downloads |
Excerpted from interview with researcher at the 2008 National Hispanic Science Network on Drug Abuse Conference in Bethesda, MD.
Conference dissemination services provided by Interlink.
More About "Research Theory, Design, and Methodology"
|
More From Liliane Windsor (bio)
|