Qualitative vs Quantitative Data Explained

Quantitative data is numerical information you can measure and do arithmetic with, while qualitative data describes qualities or categories using words and labels. This distinction is one of the first things every analyst learns because it decides which analysis methods you are allowed to use. Averaging a set of prices makes sense; averaging a list of customer cities does not. Knowing the type before you touch the data prevents a whole class of embarrassing mistakes.

The two kinds appear side by side in almost every dataset, so you rarely choose between them — you handle both at once. This page sits alongside the rest of the data analytics fundamentals, and it pairs closely with the idea of measurement levels.

Quantitative data: numbers you can compute with

Quantitative data answers "how much" or "how many." Age, height, temperature, revenue, order count, page load time — all are numeric and support mathematical operations like sum, average, minimum and maximum. Quantitative data splits into two subtypes:

• Discrete data counts whole units: number of orders, number of employees. You cannot have 3.5 orders.
• Continuous data measures on a scale and can take any value in a range: height, weight, temperature, elapsed time.

Because quantitative data supports arithmetic, it unlocks the full toolbox of statistics — means, standard deviations, correlations and forecasts.

Qualitative data: qualities and categories

Qualitative data, also called categorical data, describes attributes rather than amounts. Eye color, product category, job title, country, and open-ended survey comments are all qualitative. You cannot average them, but you can count them, find the most common value, and group other measures by them.

A crucial warning: numeric-looking values are not always quantitative. A phone number, a postal code, or a customer ID is written with digits but is really a label — averaging phone numbers is nonsense. The test is not "does it contain digits" but "does arithmetic on it make sense."

A worked example

This snippet separates the two data types in a small table and applies the right operation to each.

import pandas as pd

# illustrative customer sample data
df = pd.DataFrame({
    "customer_id": [1, 2, 3, 4],
    "city": ["Pune", "Pune", "Delhi", "Pune"],
    "age": [24, 31, 28, 45],
    "plan": ["Basic", "Pro", "Basic", "Pro"],
})

# quantitative: average makes sense
print("Average age:", df["age"].mean())

# qualitative: count categories instead
print(df["city"].value_counts())
print(df["plan"].value_counts())

Expected output:

Average age: 32.0
Pune     3
Delhi    1
Name: city, dtype: int64
Basic    2
Pro      2
Name: plan, dtype: int64

Notice the methods differ by type. For age (quantitative) we take a mean. For city and plan (qualitative) an average is meaningless, so we count occurrences instead. Applying the correct operation to each type is the whole point of the distinction.

Turning qualitative into quantitative

Analysts frequently convert qualitative data into numbers so it can be analyzed statistically. You cannot average the word "satisfied," but you can count how many customers chose it, or encode a survey scale from "very unsatisfied" to "very satisfied" as 1 through 5. This encoding must be done thoughtfully — treating unordered categories like colors as numbers would invent a ranking that does not exist. The levels of measurement framework tells you exactly when such encoding is valid.

How analysts use the distinction

Every dataset an analyst opens mixes both types. A sales table has quantitative columns (revenue, quantity) and qualitative ones (region, product, salesperson). The typical analytical move is to group by a qualitative column and aggregate a quantitative one — "average revenue by region," "order count by product." That single pattern powers most reporting, and it works precisely because you have correctly identified which columns are which.

Qualitative data also drives segmentation: splitting customers by category to compare their behavior. And open-ended qualitative text — reviews, support tickets — increasingly feeds text analysis, connecting to the world of unstructured data.

The distinction also shapes how you visualize data, which is a large part of an analyst's output. Quantitative data suits histograms, line charts and scatter plots that show distribution and trend. Qualitative data suits bar charts and pie charts that compare category counts. Choosing a chart that matches the data type is not cosmetic — a line chart implies continuity between points, so using one for unordered categories quietly misleads the viewer into seeing a trend that does not exist. Getting the pairing right is a small habit that separates clear communication from confusing charts.

Common mistakes

• Averaging categorical codes. Encoding cities as 1, 2, 3 and then averaging them produces a meaningless number.
• Treating IDs as quantitative. Customer IDs and postal codes are labels; summing or averaging them is a bug.
• Ignoring qualitative data entirely. Rich insight often hides in text feedback that beginners skip because it is harder to summarize.
• Forcing a scale onto unordered categories. Colors and product names have no natural order; pretending they do distorts analysis.

In interviews

Interviewers commonly ask you to classify data: "Is a rating from 1 to 5 qualitative or quantitative?" or "Give three examples of qualitative data." The tricky cases — phone numbers, zip codes, ratings — are where they probe understanding. A strong answer explains the test (does arithmetic make sense) rather than just guessing, and connects the distinction to which analysis methods each type allows.

Where this fits in your learning path

This distinction underpins much of the data analytics fundamentals track. It leads naturally into levels of measurement, which refines qualitative and quantitative into four more precise categories, and into structured vs unstructured data, which looks at how data is stored rather than what it measures. Together these three pages give you a complete vocabulary for describing any dataset you meet.