Qualitative and Quantitative Data Explained
Since data can take various forms, understanding the differences and intersections of qualitative and quantitative data can help you produce data-informed insights.
Qualitative data are data that have no numerical value but rather represent categories or concepts. For example, someone's affiliation to the university (student, staff, faculty) would be considered qualitative information. Though mathematical calculations cannot be used on qualitative data, descriptive statistics such as frequency can describe qualitative data. Nominal data and ordinal data are the two qualitative data types.
Nominal Data
Nominal data are data that represent categories and have no quantitative (numeric) value or natural ordering. For example, someone's nationality would be considered nominal data: American, British, French, etc. This data type can be described statistically in terms of frequency. Other examples of nominal data include hair color, religion, sexual orientation, college major, and preferred cell phone carrier.
Ordinal Data
Ordinal data are data that can be categorized and has a natural ordering but it has no quantitative (numeric) value. For example, someone's final place in a race would be considered ordinal data: first, second, third, etc. This data type can be described statistically in terms of frequency. Other examples of ordinal data include academic grades (A,B,C,D,F), education degree level (Bachelor's, Master's, Doctoral), and satisfaction rating (extremely dissatisfied, dissatisfied, neutral, satisfied, extremely satisfied).
Qualitative data are data that have numerical value and can be processed using statistical methods. For example, someone's shoe size would be considered qualitative information. Qualitative data can be further divided into discrete data, continuous interval data, and continuous ratio data.
Discrete Data
Discrete data are data that has quantitative (numeric) value and can be counted, not measured. For example, the total number of cars parked on campus would be considered discrete data. This data type can be described statistically in aggregate -- as a single number, for example, there may be 50 cars parked on campus. This aggregate number can be subdivided based on a categorical feature -- there are 10 blue cars, 30 red cars, and 10 green cars. You can run more advanced statistical tests on discrete data if you observe the data over time. With this longitudinal view, you may ask the average cars parked on campus over a given temporal period (day, week, etc.). Other examples of discrete data include the number of students in a classroom, the number of computers in a given space, and the number of light fixtures in a commercial property.
Continuous Data
Continuous data are data that have quantitative (numeric) value but are measured instead of counted. For example, someone's height would be considered continuous data. When continuous data is observed and collected in a dataset, statistical procedures can be used on the determine the mean, mode, range, standard deviation, and other statistical characteristics of the data. Other examples of continuous data include temperature in a given space, time taken to complete a task, and length of a film. Continuous data can be converted without changing the value itself --> 3 feet is the same as 1 yard.
Continuous data can also be further divided into interval and ratio data. The difference between these types is that interval data can be represented by values less than zero (temperature, for example) while ratio data cannot. Temperature is an example of interval data whereas height is an example of ratio data.
Data Type |
Description | Examples |
Nominal | can be categorized and have no quantitative (numeric) value or natural ordering | gender categories, hair color, computer manufacturer |
Ordinal | can be categorized and has a natural ordering but it has no quantitative (numeric) value | age, placement in a competition, satisfaction rating |
Discrete | has quantitative (numeric) value and can be counted, not measured. | number of cars on campus, products sold in a month, number of players on a team |
Continuous | have quantitative (numeric) value but is measured instead of counted | length of a film, body temperature, weight of a flamingo |
Python is a powerful programming language used for data analysis, web development, application creation, and other technology/data heavy tasks. Data in python can take the following forms (list not exhaustive):
Basic Data Types
Integer - numeric type presented as a whole number
Float - numeric type presented as a decimal number
String - text type including single characters
Boolean - binary type
Data Structures
List - 0 or more items stored as one object
Dictionary - 0 or more items mapped to one another in a key-attribute structure
Set - 0 or more items stored as one object, items in a set cannot be changed but new items can be added
Tuple - 0 or more items stored as one object, items in a tuple cannot be changed and no new items can be added
Data Frames (used in the pandas library) - structured data presented as rows and columns (similar to a spreadsheet or csv file). To learn more about the pandas dictionary, visit the following link: Pandas DataFrames (w3schools.com)
R is another programming language used for data-intensive tasks. Data in R can take the following forms referred to as R-objects (list not exhaustive):
Basic Data Types
Logical - binary data type (TRUE or FALSE)
Numeric - numerals inclusive of numerals with decimals
Integer - numerals exclusive of numbers with decimals (must include capital "L" after number to store value as an integer)
Complex - imaginary numerals, lowercase "i" is used to denote an imaginary number
Character - text type used for single characters and strings
Data Structures
Data structures contain 0 or more data instances.
Vectors - store multiple variables as one object
Lists - store multiple variables, including vectors, as one object
Matrices - store multiple variables in a two dimensional structure of columns and rows
Arrays - store multiple variables in any number of dimensions
Data Frames - tabular data objects consisting of columns and rows