Analysis of Qualitative Research Methods

Introduction
There are a number of different qualitative research techniques that can be used to analyze data. These include crisp and fuzzy set qualitative comparative analysis, content analysis, exploratory design and explanatory design. This paper will examine each of these individually, and seek to contrast them. Then, the best types of research for each design will be examined, and the paper will conclude with an explanation of how technology has changed qualitative research.
Qualitative Comparative Analysis
Qualitative comparative analysis is sometimes used in business settings. This is when contextual information and cognitive knowledge are blended. The data sets can either be crisp or fuzzy. The difference is that crisp data sets would be, for example, survey responses where there are a finite number of responses, and they are set out in the questions. A fuzzy set would be the output of a survey with open-ended questions. For the purposes of analysis, the big difference is that the fuzzy set you might have different responses that basically say the same thing. To properly analyze that data, the fuzzy set would have to be codified. An example would be a simple geographic response. The question “what country are you from?” with a dropdown menu would yield only United States. But if the question is open-ended, you’d get “US”, “USA”, “United States”, “United States of America” and maybe some others. So different responses that say the exact same thing.
This is a simple example. For more complex examples, it will be more difficult to code, but the principle is the same. The researcher would look for answers that are similar or the same, group them together in order to make comparisons about the data.
There are several advantages to this qualitative technique. First, it can handle both crisp and fuzzy data sets, which is the required degree of flexibility in qualitative analysis. Furthermore, This technique provides the opportunity to compare similar phenomena. Comparative analysis can help researchers draw conclusions. This technique probably works best with smaller data sets, because the comparison is often manual. However, for larger responses, the inherent flexibility of this methodology allows for comparison to still take place, but limits the utility of trying to do this over larger data sets.
Content Analysis
Content analysis is a means by which qualitative data is distilled into key pieces of information that can then be analyzed. An example might be Harrison (2013), who examined over 2000 articles to determine different types of mixed methods research. By analyzing the content of the different articles, searching for specific themes or topics, the research has the ability to determine how the information is coded, and that provides the researcher with significant flexibility. While there is a manual element to this, it is clear that the researcher also can benefit from this amount of freedom in shaping his or her own research.
The content analysis is best used for things like research where there are key content themes. It works best when there is a common thread – in the Harrison case these are all research papers. Where there is no common thread, a content analysis will simply yield that there’s nothing in common between the content in the different pieces.
The content analysis can be done over large data sets, but because it requires manual codification this process will be time-consuming. For business, it is perhaps not the best system, in the sense that content analysis would need to be automated in order to be efficient, and It often is not that.
Exploratory Design
An exploratory design is one that starts with qualitative data, leading to a quantitative follow-up. The exploratory nature uses the open-ended, fuzzy nature of qualitative data to explore a concept with the objective of being able to formulate a testable hypothesis. This type of design seeks to take a fairly broad subject, learn more about it, analyze the large data set gathered, and then to arrive at more refined knowledge that can be used to create a subsequent quantitative study. Typically, this is done for topics that are either emerging or relatively unexplored. With such concepts, the researcher might suspect that there are things of value worth exploring, but know that the knowledge level is quite large, and the gaps in the research are substantial. The exploratory nature thus serves to define future research more than anything else.
The benefit of this approach is clearly that it allows for researchers to learn more about a subject, and see if there are any threads of research that should be explored. The downside of this type of approach is that it is poorly suited for areas where there is a lot of knowledge already. Where prior research has clearly identified narrow threads of research to pursue or specific research questions that need to be answered, the exploratory approach casts too broad a net – the researcher should focus on a different methodology with a particular hypothesis to test.
Explanatory Design
An explanatory research design is one that begins with qualitative research, then there is a quantitative follow-up, and then qualitative contextual information is used to make sense of the outputs of the quantitative data. In such a situation, the first round of qualitative seems to set out the path for a quantitative hypothesis. When that hypothesis is tested, the findings might still require a degree of contextualization that qualitative can help with. This is where the resesarcher then turns back to the original qualitative data to help understand what the quantitative outputs are saying.
Explanatory design is well-suited for situations where the exploratory phase has already been completed, and there are specific research questions. With this type of research, the key is that you have some quantitative data, but that data needs some explanatory context. If the explanation can be tested quantitatively, it should, but it if cannot then that is the time for the qualitative explanatory angle.
Technology
Technology has transformed qualitative analysis because technological applications, including artificial intelligence, have made it possible to automate some qualitative data analysis. AI in particular has the ability to take similar words and instantly recognize them as being the same. It also can identify common themes, and the more complex AI and machine learning get, the more refined it will become with respect to the ability to understand the nuances of language, in in turn qualitative data. The application of AI to qualitative analysis is a well-established field at this point, and the sophistication of capabilities, if it has not transformed qualitative analysis already, will in the near future (Papagelis & Plexousakis, 2005).
Another area where technology is improving qualitative research is in the information gathering and workflow aspect. Smartphones, tablets, and their applications can help researchers to conduct surveys and process data, and then help researchers to disseminate it. A researcher with a basic understanding of the technology can perform all of these functions more quickly and without barriers. Tools that help researchers in this way can streamline the process and save themselves considerable amounts of time in their research processes (Moylan, Derr & Lindhorst, 2016).
Conclusion
There are a number of different qualitative research methods. It is important for a researcher to have a strong backing not just in the different research methods, but in the uses of those methods, and the subject matter being studied. Knowing what sort of gaps there might be in the research is an important element that can help the researcher match the right qualitative research method with the objective of the study and the types of data that can be gathered. Fortunately, technology has made the gathering and processing of qualitative data much easier than it once was, and this has in turn saved researchers substantial amounts of time.


References
Harrison, R. (2013) Using mixed methods designs in the Journal of Business Research. Journal of Business Research. Vol. 66 (2013) 2153-2162.
Moylan, C., Derr, A. & Lindhorst, T. (2016) Increasingly mobile: how new technologies can enhance qualitative research. Qualitative Social Work. Vol. 14 (1) 36.47.
Papagelis, M. & Plexousakis, D. (2005) Qualitative analysis of user and item-based prediction algorithms for recommendation agents. Institute for Computer Science. Retrieved June 1, 2018 from http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.74.2796&rep=rep1&type=pdf
Rois-Tierno, N., Huarng, K. & Ribeiro-Soriano, D. (2016) Qualitative comparative analysis: Crisp and fuzzy sets in business and management. Journal of Business Research. Vol. 69 (2016) 1261-1264.