Qualitative Research – Unlike quant

Qualitative Research – Unlike quantitative research, qualitative research is typically unstructured and exploratory in nature. In this case, the researcher is not interested in determining objective statistical conclusions or in testing a hypothesis, but rather in gaining insights about a certain topic. Common qualitative research techniques include focus groups, interviews, and observation.

Since the data is unstructured–imagine a bunch of handwritten notes from a focus group meeting–it can be tricky drawing conclusions and presenting the findings. In the case of interviews and focus groups, the moderator may simply take some time to write up the key points heard in the meeting, and then present those key points to the interested parties. For example, in a focus group about pizza, you might see the following summary: “common concerns among partipants were cheese overuse, greasiness, and bland sauce.”

Another approach when it comes time for qualitative analysis is to “code” the unstructured data, in an attempt to form the data into something that can be summarized with tables or charts. If the researcher conducted 20 interviews and asked similar questions to each person, responses might be summarized, or “coded,” into short descriptions. A coded response to the question “when do you wear a watch?” might be something like “3 – formal situations.” You can imaging other answer codes might be “1 – never,” “2 – everyday,” etc. With the conversations summarized into these coded responses, the data has been converted from purely qualitative data into quantitative data that can be summarized in charts and graphs.

Yet another qualitative analysis method is automated content analysis. Let’s say you have a big heap of unstructured text that you’ve typed up during a focus group. You could manually look through the notes and draw some conclusions. You could also take that text and dump it into a content analyzer (e.g. wordle), that will look for word frequency and kick out a nifty “word cloud” of the key words being used. This method provides a quick way to gain insights into the unstructured data, especially when the set of data is overwhelmingly large.

Example: Let’s look at an example of qualitative research from start to finish. Imagine you work at a bowling alley and you have a little form at each lane that asked “how was your experience today?” Let’s assume there are 100 responses, including comments like “great, the staff was so courteous” and “terrible, I’ve never bowled so poorly in my life.” When the boss asks, “how is the feedback so far?” you stratch your head because you are not quite sure how to summarize the data. You have a few choices here. (1) You could take the pile of 100 cards and dump them on your boss’ desk and say, “here, why don’t you look through them yourself?” Some managers will be OK with this response. (2) You could look through the cards, notice some key themes and tell your boss, “most of the feedback is positive, but there were several comments about the bathroom being too dirty.” (3) You could go a step further and code each response into a spreadsheet, perhaps classifying each response as positive, negative, or neutral, and perhaps into sub-categories like “food,” “cleanliness,” “staff,” or whatever makes sense. They you could return to your boss and say something like “65% of responses were positive, with 1/3 of those mentioning the food.”