Units of Analysis – Week 1

 

 

Hi all,

Regarding units of analysis, there was a question on my quiz that had me doing some research. De Vaus writes that, "Any region (e.g. country, county, state) could be used", and I've found one site that says, "it is the analysis you do in your study that determines what the unit is."

Based on this, to determine the unit of analysis, wouldn't you need to know either; A. Something about the conclusions of the research (as is the case for most of Week 1's questions on units of analysis), or B. The type of analysis being carried out by the researcher? And if neither of these are present, could one define the cases/UoA and variables in different ways?

Anyone is welcome to contribute to this if they have some thoughts on the topic.

Thanks,

 

 

My thoughts are that I think about the unit of analysis first and foremost in terms of the research question. Also can be determined in terms of the conclusions and possibly type of analysis (depending on how well it is spelled out, you should be able to, but the descriptions in research papers are not always framed clearly enough). The unit then changes according to your question (and the analysis flows from that), using data in the same file. Collecting data on number of abortions for women aged 20-24 in the UK for every year from 1968-1998 where the question is "how many abortions per 1000 women aged 20-24 in 1998?" makes women aged 20-24 the unit of analysis, using the same data set and asking "has the abortion rate for women aged 20-24 changed between 1968 and 1998?" flips the unit of analysis to year.

 

You seem to have highlighted the research question, the conclusions and the type of analysis as the key points in determining what the unit of analysis is. I agree.

And I would add that I'm starting to believe the unit of analysis is impossible to determine if none of these are mentioned. Which might mean that one of the quiz questions has two possible answers.

Regards,