When analyzing data like age groups, salary ranges, or grade levels, it’s often not meaningful to analyze individual values. A more practical approach is to group continuous data into intervals, and then perform analysis on those groups.
This article introduces the basic usage of pd.cut() and how to combine it with groupby() for categorized statistical analysis.
Basic Usage of cut
#
Suppose we have student score data, and we want to categorize the scores into three levels: “Fail”, “Pass”, and “Excellent”, and then calculate the average score in each level.
Because scores are continuous, using groupby() directly would generate too many groups. Instead, we use binning to divide the scores into a few defined intervals using pd.cut():
df = pd.DataFrame({
'Name': ['Ming', 'Mei', 'Qiang', 'An', 'Jie'],
'Score': [55, 70, 82, 88, 93]
})
We want to divide them as:
- 0–60: Fail
- 60–80: Pass
- 80–100: Excellent
df['Level'] = pd.cut(
df['Score'],
bins=[0, 60, 80, 100],
labels=['Fail', 'Pass', 'Excellent'],
right=True
)
Parameter Explanation #
| Parameter | Description |
|---|---|
| First argument | The numeric Series to bin |
bins |
The interval edges; can be an integer (equal-width bins) or a list |
right |
Whether to include the right edge (default is True → (a, b]) |
labels |
Optional labels for each interval |
include_lowest |
Whether to include the lowest value in the first interval |
Note: If a value exceeds the maximum bin (e.g., Score > 100), the result will be
NaN.
Result:
Name Score Level
0 Ming 55 Fail
1 Mei 70 Pass
2 Qiang 82 Excellent
3 An 88 Excellent
4 Jie 93 Excellent
Combine with groupby for Analysis
#
Once grouped, we can use groupby() to analyze each category.
For example, calculate the average score for each level:
df.groupby("Level")["Score"].mean()
Output:
Level
Fail 55.000000
Pass 70.000000
Excellent 87.666667
Name: Score, dtype: float64
By binning first and then analyzing, you can organize your data exploration in a clearer, more structured way.