What is a unit of analysis in qualitative research?

I get confused by Units of analysis and units of observations. I'm doing a ethnographic case study of after-school programs that include some form of the arts or arts education. I want to learn what are the key charecteristics such programs need to have in order to be collaborativly supportive of arts education standards and benchmarks of the school (district). I THINK my the after-school programs themselves are the units of observation. The units of analysis would be what the various actors in these partnerships have to say (i.e - their documents or social artifacts) about arts education standards and collaborating with schools.

What is a unit of analysis in qualitative research?
I would say that your unit depends on how your study is designed. My background is not in education, but in statistics. I could not find a definition of this specific term on the net. So I am deconstructing it.





Your unit is what you are counting by. One item is a unit. So what are you using for your measurement. Are you looking at classrooms, schools, school districts, any of these could be your unit.





So look at your design and ask, what EXACTLY am I observing, what EXACTLY am I analyzing?





You say that you are studying documents or artifacts, so from what you wrote, I would say that your unit would be the number of artifacts you study, since you can't include programs in your study which do not involve artifacts.








But if the benchmarks are student test scores, I would say your unit of observation is the number of students.





The best bet is to take a study of similar design, not necessarily studying the same subject, but a study designed in the same basic manner. Make sure that the study is one that is recognized as being well-designed.





Use that study and report as a blueprint for your own report, just like you might do for a contract, or a job application. I am not telling you to copy anything, because the material will be what you found in your own research.





Ask yourself what that study uses as a unit of observation, and unit of analysis. If you cannot find a comparable element in your own study, then you may have a poorly designed study.





Some researchers confuse their subject matter with the design of the study. They also sometimes KNOW what they want to look for, and confuse this HOW they will look for it. The two are related but not the same.





One of the main purposes of research is to try to minimize the observer's bias. So with a properly designed study, you should be able to hand your study to the cashier in the cafeteria, and say, Here, do this study, and they should come back with the same results.

dog teeth