Table of Contents
What is Qualitative Data?
Qualitative data is the information set that contains words, subjects, descriptions, definitions, observations, and categories without any numeric data. In a way, you can assume it is the opposite of quantitative data. A comparison between these two concepts and examples would be helpful for better understanding which is critical for data visualization.
Qualitative vs Quantitative Data
Before comparing qualitative data and quantitative data, we have to say what quantitative data is. Simply, it is data that uses mostly numbers to define and describe the content. Lately, because qualitative research became popular, the following comparison is trendy.
Quantitative data answers what, when, where, and how much/many,
Qualitative data answers who, how, and why
As you can understand, the first one is about questions that can have very short answers which is easy to categorize. On the other hand, answers to the questions of the second data type are way longer and descriptive. So, it is hard to categorize them easily. I am using the “categorization” word for a reason. It is important for the visualization of data whether it is qualitative or quantitative. Because most of the visualization tools are focusing on the categories. We will talk about that later in this article.
Qualitative Data Examples
As a lifelong learner, examples are the most efficient part of validating the lessons learned. Let’s go through the examples of qualitative data.
The first example is about our beloved coffee. Check this conversation:
You buy a cup of coffee and tell your friends about it.
Hey, this time I bought a medium cup of Ethiopian coffee thanks to the barista. The flavour is inimitable, sensitive and delicate; I can sense notes of jasmine flower, bergamot and blueberry.
As you can see, there are lots of information about coffee but nothing is quantitative. This time let’s check the quantitative version.
Hey, I bought 14 oz cup of Ethiopian coffee grown at 2300 m. It is so delicious I can sense at least 3 flavor notes.
Now, you can easily identify the difference I assume. Think you have a collection of sentences like the above ones. How would you visualize it? Of course, we have to analyze it first. We will go through that but before, let’s check some real data set.
Even the name of the collection will ring the bell. “Study Number 2000 – Family Life and Work Experience Before 1918, 1870-1973”
This is a collection of interview documents. As a researcher, you are looking for information regarding health in the UK in the late 1800s. Luckily, each document contains a section called “Health and childbirth”. Here are a couple of samples;
Health and childbirth
Parents paid into Ancient Order of Foresters and doctor’s club. Mother had severe headaches. Mother attended by local woman but doctor also present at births.
Health and childbirth
Eye problems necessitated hospital visits. Broken ankle in tractor accident which kept him out of the army. Refers to himself as a cripple.
Health and childbirth
Husband injured as chimney sweep, ill for three and a half years. TB test at Colindale Tubercular Hospital negative though convinced he had TB. Born early years of marriage. Doctor present at confinements but ‘a terrible doctor’ – difficult birth.
There is a lot of information which you cannot understand at first sight. You have to read all to extract information for your research. Since it requires a bit of hard work, researchers use a methodology/process called qualitative data analysis.
What is Qualitative Data Analysis?
The purpose of qualitative analysis is a better understanding of the insights that data has to offer. The quality, characteristics, and meanings of that domain can be considered as some of them. In that manner, the common definition of Qualitative Data Analysis (QDA) is the range of processes and procedures used on the qualitative data to transform them into some form of explanation. To me, we have to extend this explanation such as;
Qualitative Data Analysis (QDA) is the range of processes and procedures used on the qualitative data to transform them into quantitative data providing metrics that can be visualized for clarification and validation.
I do not dive deep in qualitative analysis in this post but the following steps are required to analyze the data set.
Steps of Qualitative Data Analysis (See Details)
- Prepare and organize your data.
- Review and explore the data.
- Create initial codes.
- Review those codes and revise or combine them into themes.
- Present themes in a cohesive manner.
If you need to see more in qualitative data analysis, this course is a great source.
Qualitative Data Visualization
To be honest, this part is the summary of this great post written by Lydia Hooper.
In our minds, we aim to build connections between ideas. Then, we group them into categories and start to think about their relations, etc. When I first read qualitative data samples, I immediately feel that happening. Intuitively, I dream of myself standing by a whiteboard taking notes, grouping specific words, counting them, and eventually drawing lines as connections. The best visualization tool to describe this is the Mind Map. They are great tools for analyzing the data because you can highlight relations between categories, ideas, etc.
Another option is using Word Clouds since qualitative data is mostly textual and non-numeric. If you are familiar with text mining tools or anything that can extract word counts for adjectives or any other descriptive words, you can use the results for further analysis. The below image shows the results from the whole data set that we used as a sample above.
Another effective tool is the Word Tree for the analysis. It is created with d3.js. As the previous visual, I use this great visualization tool to highlight words in the same qualitative data set. It is kind of a combination of word cloud and mind map.
This tool obviously says that you have to focus on family members and miscellaneous cases. And yes, you can click and navigate through these words with the help of this tool. You can go much deeper by using visualization tools.
Importance of Data Visualization
To sum up, visualization tools are not useful only for showing results, but they also help you broaden your viewpoint. We have a huge amount of data that hides unknown information. Even we are talking about unknown unknowns. Qualitative data is the center of this problem and we have to work on it harder.