The debate of qualitative versus quantitative data/research is never-ending. But spending time debating something that can never be definitively proven and that ultimately depends on individual preference isn’t productive, at least from my point of view. Instead, I believe our time is better spent moving the discourse to a more meaningful topic—one that is less controversial but equally important to dissect.
Quantifying qualitative data is not a new phenomenon. Personally, I have met many curious individuals asking about this topic, but I have yet to meet anyone trying to turn quantitative data into qualitative data. Notwithstanding mixed methods, where the goal is for qualitative and quantitative approaches to supplement or complement each other, I am not even sure if there’s a term for that. QUALITYFYING? It doesn’t sound right. Never mind.
Regardless, I do understand the appeal of assigning numbers to feelings, expressions, thoughts, and perceptions. It is, however, much more complex and layered than that. Before attempting to quantify qualitative data, the first question I would ask is: Is it appropriate? Does it make sense, and does it help address the research question?
In a scholarly article published all the way back in 1952, Hayashi argued that the quantification of qualitative data obtained by measurements and observations is an important method in both social and natural science research. Of course, this article was published in the journal “Annals of the Institute of Statistical Mathematics,” and Hayashi clearly states that the entire article is written from a mathematico-statistical point of view. He, assuming that is the preferred pronoun, also adds a caveat: “if the method of quantification is not reasonable.” This raises another question about the methodological appropriateness of quantification. This is an op-ed, so I will not delve too deeply into the methodological rabbit hole—that’s a whole different conversation and opens a can of worms that would divert the narrative.
Let’s circle back to the topic—quantification of qualitative data in general. As I mentioned, it really depends on the research question and whether quantifying qualitative data strengthens the research or contributes to answering the research question while maintaining methodological rigor and ethical standards.
But one can certainly entertain the question: Why even attempt to quantify qualitative data? The most common response you will hear is that it is inherently driven by an obsession with numbers. Not for everyone, of course, but for those for whom numerical data speaks louder than textual data. Economists and market researchers come to mind. However, it’s not that simple. Often, it’s not only about the researcher’s preference—what researchers think is right or appropriate—but rather what stakeholders want to see. It is not surprising to say that researchers are often put in uncomfortable positions. It is not difficult to find a researcher who has questioned their research methodology in hindsight—whatever the reason may be.
From my experience in academia, industry, and non-profit sectors, I can attest that this happens more often in certain areas, such as market research, compared to academic research, where the ultimate goal is to publish. Of course, to avoid offending fellow academics, let’s also say that their purpose is to contribute to broader theory and knowledge production. Even within sectors, it depends on the discipline.
Proponents of quantitative methods are more likely to lean towards quantification—not always, but most likely. Quantifying qualitative data allows for broader generalizability and comparative analysis, which is particularly useful for large datasets. However, this raises several important questions: What do we lose in the process of quantification? How do we ensure that the richness of the qualitative data is preserved when assigning numerical values to inherently nuanced information? The core of qualitative research lies in capturing the depth, context, and complexity of human experiences. Reducing these experiences to numbers can risk oversimplification, stripping away the meaning and context that make qualitative data so valuable in the first place.
One key question to consider is whether quantification is truly adding value. Does it help answer the research question in a more effective way, or does it merely make the findings more presentable to stakeholders who are more comfortable with numbers?
Quantification can sometimes result in losing the subtlety and diversity of qualitative insights. For instance, when we assign numerical codes to themes or count the frequency of certain responses, we might overlook the nuanced variations that are crucial for a deeper understanding. This is particularly challenging when dealing with data that involves emotions, complex behaviors, or cultural contexts, where meaning cannot be easily reduced to numbers.
Another question to ponder is the methodological appropriateness. Are we using the right tools and approaches to quantify qualitative data? Inappropriate quantification can lead to misleading conclusions, especially if the method does not align well with the nature of the data. For example, qualitative data from ethnographic observations or in-depth interviews is rich with context, making it challenging to quantify without losing critical information. Not inappropriate or useful, just more challenging.
On the other hand, survey-based qualitative data might be more straightforward to quantify, but even then, careful consideration is needed to ensure the integrity of the data.
The repercussions of quantifying qualitative data can be significant—both positive and negative. If done poorly, it can lead to misguided decisions, misinterpretation of results, and a loss of valuable insights. In fields like anthropology, where the focus is on understanding cultural nuances, quantification can be particularly problematic, not always. I can say with confidence, it will be challenging though. Conversely, in fields like economics, where there is a preference for numbers, quantification is often seen as necessary. According to Nardo (2003), data obtained from business and consumer surveys are often used in forecasting models and in testing different expectation formation schemes. Their use, however, requires a previous step of transforming the qualitative data into quantitative figures. Nardo provides a critical review of different quantification methods, highlighting the limits of their use in macroeconomic modeling, if anyone is interested in reading in detail. Notably, this paper is published in the “Journal of Economic Surveys,” reflecting the field’s emphasis on quantifiable data.
I have always been a qualitative researcher at heart. I have worked on projects where quantification of qualitative data made sense and added value, and projects where quantification did not add value. But I do believe, strongly, that ultimately the decision to quantify qualitative data should be guided by the research question and the nature of the data. It is important to ask: Does quantification help answer the research question more effectively, and does it maintain the ethical and methodological rigor required? There is no right or wrong answer. If quantifying qualitative data works for you and strengthens your research, that is great. Just be sure to do so while holding ethical standards and maintaining the rigor needed to preserve the value of the original qualitative insights.
As I conclude this opinion piece, I find myself wondering if “quantifying qualitative data” is just as controversial as “qualitative vs quantitative.” But then, I believe it can be controversial, though perhaps not to the extent of qualitative vs quantitative, primarily because there is an appropriate way to do it, and there are stringent questions one can ask beforehand, albeit in a blurred, gray area. Unlike the qualitative vs quantitative debate, the line here is much blurrier, much grayer. Or is it?
Ultimately, if quantification helps you gain insights without compromising the richness of the original data, go ahead. Just remember to always ask yourself whether it is appropriate and, most importantly, be mindful of ethical standards and methodological rigor. So, I guess don’t stop at “Does your action speak louder than your words?” but also ask “Does your number speak louder than your words?” Of course, at the right time and in the right place.