Understanding the Differences Between Quantitative and Qualitative Research

Research drives the modern economy. In order to innovate and understand what really drives consumer decisions, brands must keep their finger on the pulse of what’s going on in their industry.

There are two types of empirical research: quantitative and qualitative. But the differences between them are a little more complicated than the expression “quality vs. quantity” would suggest. 

What is quantitative research?

Quantitative research deals with facts, usually in the form of numbers. The word quantity actually comes from the Greek quantus,meaning “how much.” Most people are accustomed to hearing about opinion polls in politics and statistical data in medical research. These are both examples of quantitative research.

The defining feature of quantitative research is that the data can be collected at a large scale, due to the conformity of the questions.

There are three primary methods of collecting quantitative data: surveys, observations, and experiments.

Surveys - At some point or another, practically everyone has encountered a survey or questionnaire. When you call your bank or internet provider, often you’re encouraged to “stay on the line” after the call and rate your experience. Some of the receipts from shopping trips will ask you to go online and participate in a customer-satisfaction survey.

These are examples of quantitative research in the form of surveys. One crucial feature of such surveys is that the questions are closed-ended. In other words, your possible responses are limited to a specific range. Sometimes the range is limited to two — i.e., yes/no, check the box, and so on. Sometimes the questions are multiple-choice or provide a scale, such as using a five-star system to rate an experience or product. 

Experiments - When the goal is to establish a cause-and-effect relationship, scientists are likely to design experiments wherein certain variables are controlled. A good example of this kind of quantitative research that many people will be familiar with right now is a vaccine trial. 

Note that the results of an experiment — even a complex one — can still be expressed numerically. In a medical trial, for instance, a vaccine can be shown to be either effective or ineffective. Closer scrutiny can gauge the vaccine’s effectiveness using scales and percentages and also account for the influence of other data, such as participants’ ages.

Observations - The results of an observation study — simply observing an environment or process without controlling for variables — can also be expressed numerically, and therefore count as quantitative data. 

For example, imagine a botanist studying the range of flora in a park. Instead of counting every plant, the researcher stakes out smaller, more manageable zones where the diversity of species is tabulated. The percentages can then be extrapolated to the entire park area. 

What is Qualitative Research?

In qualitative research (from the Latin qualis, meaning “of what kind”), we aim to describe the subject of our study more than quantify, or measure, it. In other words, the results of a qualitative research study will be expressed primarily in words, not numbers. This is because we are dealing more with ideas and experiences than with hard facts.  

The purpose of this kind of research is to dive deep into the subject at hand and arrive at meaningful, rich insights that help us understand the subject in all its complexity and nuance. As a result, qualitative studies are usually much less structured than quantitative ones, and the sample sizes are far smaller.

Qualitative research usually includes one or more of the following methods: 

Open-ended questions - Once a group of respondents has been recruited, they are surveyed using open-ended questions, as opposed to the closed-ended questions used in quantitative research. Instead of a questionnaire with a rating scale of 1 to 5, a respondent might simply be interviewed and allowed to answer with their own words. Instead of a telephone survey involving hundreds of people, researchers might recruit many fewer people for a focus group and guided discussion.

Observations - Both qualitative and quantitative research projects can make use of simple objective observation as a key technique. In qualitative research, however, the goal is to describe, not measure. A common example of this technique is the ethnography, when a researcher spends a significant amount of time embedded with a group of people in order to observe their behaviors.

Literature review - To be sure, any researcher worth their salt will begin a project by surveying the available literature on their subject. Yet some qualitative projects are built exclusively on the published work of experts. By extensively reviewing the literature on a subject, a researcher can add fresh insight or spot overarching patterns or identify oversights. 

Qualitative or quantitative: which should you choose?

Researchers determine whether to use quantitative or qualitative research methods by considering their business objectives and the end purpose of the research.

If they want to understand a subject on a “big-picture” level, then having a broad scope and accumulating as much factual information as possible would be a prudent strategy. This means quantitative.  

If, on the other hand, a deeper understanding of the details and nuances behind the facts is desired, then qualitative research is more likely to be effective.

More often than not — especially in this current age of Big Data — researchers will choose a mixed-method approach. This can extremely effective as the numbers, statistics and graphs of quantitative data can provide that big-picture understanding of trends and context, while more open-ended interactions or deeper dives into the literature can help uncover the kinds of insights that often lead to innovation and breakthrough. 

What if, instead, you didn’t have to choose?

Streetbees gathers and decodes the richness of qualitative data, at the scale of quantitative, on a continuous basis. By applying natural language processing on the unstructured data gathered - photos, videos and open text - we can uncover what people want and what really matters to them. Deep neural networks are then used to cluster similar demand choices together - free from human bias - and can pick up growth opportunities and consumption trends that were previously unavailable.

By applying AI instead of manual data processing, we don’t just save time (30 seconds vs 126 hours to analyse 1,000 submissions) but the machine is able to decode the nuances of emotions in a way that people can’t.

The problem is, human brains lack the ability to conceptualise something with so many dimensions. Algorithms are needed to work with this unstructured data and reduce it down into something we can understand. Through this, Streetbees’ research has shown that elements like context and mood predict about 75% of someone’s consumption decisions, while demographic factors (like who they are) are only responsible for around 25%.

To learn more about Streetbees’ Always ON platform, and how to gain access to qualitative data at the scale of quant, get in touch with the team here.