Some Thoughts on the Distinction between Qualitative and Quantitative Research

Today I want to offer some thoughts on the distinction between qualitative and quantitative research.  This is important to me, since a lot of people seem to think that legitimate and reliable science has to be quantitative.  I do consider the qualitative vs. quantitative distinction to be one of the fundamental epistemic dimensions, which means that every principled way of knowing or coming to believe something would have to essentially be oriented toward one of these two, although I do also acknowledge that some epistemes involve a mixture of these two poles.

If an episteme involves quantities, which are things that can be counted and/or measured in some way, then it is quantitative.  On the other hand, if it involves descriptions of unique circumstances that cannot easily be quantified, then it is qualitative.  We can reliably measure a diverse array of phenomena such as space, time, mass, light frequency, brightness, wavelength, shape, location, velocity, and acceleration.  We can also simply count individual units of any type of thing that is similar, or that at least have some sort of identifiable similarity.

The most reliable forms of science make extensive use of quantification.  Science operates best with numbers because this allows us to converge the senses, any of which can deceive us.  We can, for example, see, touch, and hear the same quantification, which is the same numeric representation of the quantity of the phenomenon that is going on.  Scientific research works best when it is driven by numbers, but this is not because of any overarching effort to reduce the world to quantity, nor to eliminate any genuine qualitative distinctions and categories.  Rather, this is because quantification helps us to develop, foster, and sustain the trustworthiness of our information gathering.  Numbers allow us to reduce the personal noise that we might be introducing into our observations and to more accurately focus on what we are modeling so that our understanding is driven by mind-independent facts rather than by self-deception.

Quantification might seem quite natural to us, but based on certain perspectives, nothing is ever exactly the same.  In the real world, things are so often unique and complex and mixed up with all kinds of things, some of which are similar.  At a basic level, we can understand and describe each moment, each place in the world, and the quality of everything as it is and as it changes.  We can recognize similarities and classify them and differentiate and organize and categorize them.  We can notice cross-similarities and differences and correlations.  We can identify specific substances, entities, events, causes, structures.  Qualitative research is essential and often involves categorization, association, and recursive hierarchical sorting, and the identification of relations, properties, wholes and parts, and essential attributes.  We need to do this in order to count similar things and in order to measure things.  Thus qualification is often necessary for quantification.

In truth, all circumstances in reality are unique in their own right, but we try to find generalizations among the particular circumstances that seem similar in certain ways.  We have to first make categorical generalizations amid the complexity and ubiquitous uniqueness of the world in order for anything similar to be counted and before any measurements can take place.  Every point in space and every instant in time is unique and this means that if we try to measure space or time, we are introducing a generalization to these unique circumstances.

In any observed phenomena, lots of things might be moving and constantly changing and the first thing that a researcher needs to figure out is what are the different kinds of things and what is changing over time and what is staying the same and what are the relations to other things.  Somewhere down the line, these things might become countable and measurable.  Similar sorts of things or dimensions of some sort can be discerned and then quantities can be computed.

Note that understanding this distinction does not imply that reality must be quantifiable nor that we cannot know something unless it is quantifiable nor that all empirical knowledge must be quantifiable.  Sometimes it happens that detailed unique descriptions must be gathered and from there it might be possible to find certain types of generalizations within this data, which could then be quantified.  Thus it is silly that some people think that science has to be quantitative.

Usually, quantitative data is objective, but there are circumstances where intersubjective data can be quantifiable as well.  This would probably include aspects of color such as brightness and saturation, the volume and tone of sounds, and degrees of pain, among other phenomena.  For anything that is intersubjective and quantifiable, it is very difficult to measure with clearly understood units and the measurements will probably always be rough estimates, and thus social verification is difficult, although this is easier if the goal is merely to rank and compare different subjective experiences that one might feel within their own body or perhaps between oneself and one or more others.  If the only goal is to figure out which experience is higher or lower or better or worse, then that is more doable.  We can know what bad pain is in comparison to not-so-bad pain and also in comparison to really bad pain.  Comparisons such as these rely on a certain kind of rough quantification that can be mutually understood to a large extent.

The quantitative vs. qualitative distinction is closely related to the distinction between so-called hard and soft sciences.  This distinction hinges on whether findings and conclusions are truly reproducible or are dependent on particular circumstances and complexities that cannot be reliably reproduced and studied with the highest level of clarity and mutual understandability.  Physics and chemistry are considered hard sciences and biology is usually considered hard as well, although it sometimes extends into territory that is a bit soft.  Psychology can be both hard and soft, depending on the circumstances and what aspects of the mind and of animal or human behavior are being studied.  The social sciences, including sociology, anthropology, linguistics, economics, and political science, are very much in the realm of soft science because they necessarily operate within complex environments and nearly all research projects will require a significant amount of interpretation of qualitative data rather than simply making measurements and cold calculations.  This fact, however, does not de-legitimize any soft science because there is a huge difference between evidence-based and peer-reviewed qualitative research and pseudoscience.