Hi, I am doing the pg major project and considering using content analysis as the research method. Since I have three different research questions, I am wondering whether the sample of each research question must be the same.
For instance, could I use different sampling methods to collect data for different questions?
Looking forward to your reply. Thank you very much.
Xi - content analysis is not your research method or methodology. It is part of your research design i.e. sampling, ethics, data collection and data analysis. If you are using content analysis to analyse your data - one assumes that you are using a qualitative methodology. If this is the case, then you would only collect one data set from your one chosen sample; therefor your sampling method will be the same for all of your intended population.
The two attached chapters may assist on qualitative sampling and data analysis.
The following sources may further help, particularly: Chapter 6 Sampling (Krippendorf, 2004, pp. 111-124).
- Elo, S. and Kyngäs, H. (2008). The qualitative content analysis process. Journal of advanced nursing, 62, 1, pp. 107-115.
- Krippendorf, K. (2004). Content Analysis: An Introduction to its Methodology, London, Sage Publications, Inc.
- Morgan, D. L. (1993). Qualitative Content Analysis: A Guide to Paths Not Taken. Qualitative Health Research, 3, 1, pp. 112-121.
Yes, statistical sampling is not an issue in qualitative methodology. But still, if you have too many sources for practical content analysis, you may need to use a representative sample, such as every 10th source. You probably can't do content analysis on 10,000 source documents, for example.
You don't really say if your are really doing qualitative analysis, or if you are quantifying your content analysis data for statistical analysls.
If you are doing the latter, I would use the SAME sample for every question. The sampling says "these sources are representative of the whole for purposes of this study." I think it would confuse things if you re-sample for every question.
A lot depends on what your goals are for your particular study, as well as the nature of the data available. For example, if you have a large number of sources available, and you want to have a representative sample, then random sampling could be reasonable.
Alternatively, from a qualitative point of view, you might want to pursue purposive sampling, where you select your cases according to their ability to meet the purposes of your research. This would be relevant if your three different research questions each require different types of sources in order to meet your research goals.
But if same item (story or whatever) could address any of your research questions, then you would not need three different samples. Instead, you would just code whatever material was relevant to your analyses within each story.
It depends on your research design. But when your question at hand regards a certain population why you should need different samples? unless you mean you need a control team, but this implies experimental research design. I propose to search and read "Qualitative Text Analysis" by Udo Kuckartz. Its a brief publication from SAGE books only 160 pages. By default when doing qualitative research the sampling ends when you reach a "point" of saturation where no more new categories and concepts emerges from your sample data. Then your sample is complete and representative. You can also watch on you tube https://www.youtube.com/watch?v=wbdN_sLWl88
Hello all! I have a question which is pertinent to this post. When it comes to word counting in a content analysis of academic papers, do we also include words that occur in tables, figure captions, table captions and title sections of a paper in case these belong to one of our categories?
I am looking forward to your reply!
hi,what is difference between content analysis and thematic analysis?
looking forward to your reply.thank you very much
Plz see this article. Hope it will fullfil your need.
Content analysis and thematic analysis: Implications for
conducting a qualitative descriptive study by Mojtaba Vaismoradi, Hannele Turunen, and Terese Bondas