In statistics, data is classified into two main categories: qualitative and quantitative. These types determine how data can be analyzed and the statistical or graphical tools that are suitable for them. Here is a detailed explanation:
- Qualitative (or categorical) data
Qualitative data describe qualities, characteristics or categories that are not measurable by numbers in a natural way. They are based on attributes or labels.
Features:
- They cannot be added or subtracted directly.
- They are often expressed by words or codes (e.g., "yes/no", "red/blue").
- They are divided into two subtypes:
- Nominal : No order or hierarchy between categories.
- Examples: Colors (red, blue, green), genders (masculine, feminine), fruit types (apple, banana).
- Ordinales : Categories have an order or ranking, but the differences between them are not necessarily measurable.
- Examples: Satisfaction levels (low, medium, high), clothing sizes (S, M, L).
Analysis and tools:
- Measurements: Frequencies (e.g. 20% of "yes"), mode (most frequent category).
- Graphs: Bar charts, pie charts (pie charts).
- Example: In a survey, 40% of respondents prefer tea, 60% coffee (nominal data).
- Quantitative (or numerical) data
Quantitative data represent quantities measurable by numbers. They allow mathematical operations like addition or multiplication.
Features:
- They express a magnitude or measure.
- They are divided into two subtypes:
- Discrete : Integer values, often from counting, with a finite or countable number of options.
- Examples: Number of children in a family (0, 1, 2...), number of cars sold (5, 10, 15).
- Continues : Values that can take any real in a range, often from measurements.
- Examples: Person’s height (1.75 m), temperature (23.4 °C), running time (12.56 s).
Analysis and tools:
- Measurements: Mean, median, standard deviation, range, quartiles.
- Graphics: Histograms (discrete or continuous), box whiskers, point clouds.
- Example: The average ages in a group is 35 years (discrete data), or the average temperature is 22.5 °C (continuous data).
Key differences
| Criterion | Qualitative | Quantitative |
| Nature | Categories, labels | Measurable numbers |
|
Examples |
Color, opinion | Weight, age, income |
| Operations | No addition or subtraction | Addition, average, etc. |
|
Subtypes |
Nominales, ordinales | Discrete, continue |
| Graphics | Bars, camemberts | Histograms, boxplots |
Why is this distinction important?
- Analysis methods : Qualitative data require counts or percentages, while quantitative ones allow for more complex calculations (mean, variance).
- Interpretation : Ordinal data (e.g., "good/average/bad") cannot be averaged as quantitative data (e.g., grades from 0 to 20).
- Context : The choice of tools depends on the type. For example, a histogram is not suitable for nominal data like colors.
In summary, let’s say that qualitative data describe "what" or "which type", while quantitative measures "how much" or "to what extent". This classification is the basis of any statistical analysis!