Hello,
I am writing my bachelor thesis right now and am conducting a quantitative context analysis looking at how terrorism is portrayed in newspapers, but certain words could be used to describe both the perpetrator and the victims. In a qualitative content analysis you have more freedom to read "between the lines" of the codes, but that is not what my issue is. I would like to know if it's possible in a quantitative context analysis, to look at the oblivious context of the code, meaning, not interpreting what they "could mean". I know quantitative content analysis means that you are looking at the occurrence of words, but I would like to know to what extent you are able to look at the context of the codes?
I cannot find anything about this in the books I've read (Neuendorff, 2017; Krippendorff, 2004), so if anyone knows another source that could help me, or know that I missed this in the books, I'd appreciate any help! :)
Hello Carl,
Here is the best resource for your question.
Conducting Qualitative Research on Terrorism: Finding and Using Primary Sources
https://methods.sagepub.com/case/conducting-qualitative-research-on-terrorism-finding-using-primary-sources
Cheers and be safe!
I am not quite sure what you are asking as "what the words could mean" is always an issue, even in content analysis. The use of a codebook generated by SMEs (journalism professionals or terrorism readers of generalism) could help you generate initial definitions that are inclusive and provide the initial definitional instructions that meet the parameters that are of interest to you. Good training for your coders (who should be selected from the population that results are going to be generalized) will help generate meaningful quantitative data. BTW, unitization of the text may be key here. Be sure to have multiple coders to provide the takeaway you are intending and is also reliable. -jd
It sounds you may be doing an "automated" content analysis, where you are asking the software to locate and count the content for you. If so, the possibility that the same content could be associated with either the perpetrator or the victims would indeed be a problem.
If so, I do not see anyway around reading the relevant text to determine the appropriate coding. In particular, you might examine the relevant words using a "Key Words in Context" approach.
- Badges
- Science topic
- Similar topics
- Books