describe() for statistical summary in Pandas - Cheat Sheet & Quick Revision

Recall & Review

beginner

What does the describe() function in pandas do?

It provides a quick statistical summary of numerical columns in a DataFrame, including count, mean, standard deviation, min, max, and quartiles.

beginner

Which statistics are included by default when you use describe() on a DataFrame?

Count, mean, standard deviation (std), minimum (min), 25% percentile, 50% percentile (median), 75% percentile, and maximum (max).

intermediate

How can you get a statistical summary of categorical data using describe()?

Use describe(include='object') or describe(include='category') to get count, unique values, top (most frequent) value, and frequency of the top value.

beginner

What is the output type of describe()?

It returns a pandas DataFrame containing the summary statistics for each column.

beginner

Can describe() be used on a single pandas Series? What does it return?

Yes, it can be used on a Series and returns a Series with summary statistics similar to those for DataFrames.

What does df.describe() show by default for numerical columns?

A Count, unique, top, freq

B Only mean and median

C Count, mean, std, min, 25%, 50%, 75%, max

D Only min and max

By default, describe() shows count, mean, standard deviation, min, quartiles, and max for numerical columns.

How do you get a summary of categorical columns using describe()?

A Use <code>describe(numeric_only=True)</code>

B Use <code>describe(include='object')</code>

C Use <code>describe()</code> without arguments

D Use <code>describe(exclude='object')</code>

To summarize categorical data, specify include='object' or include='category' in describe().

What type of object does describe() return?

A A pandas DataFrame or Series

B A Python list

C A NumPy array

D A string summary

describe() returns a pandas DataFrame when called on a DataFrame, or a Series when called on a Series.

Which of these is NOT included in the default describe() output for numerical data?

A Median

B Standard deviation

C Maximum

D Mode

Mode is not included by default in describe() output for numerical data.

If you want to include all columns regardless of type in describe(), what should you do?

A Use <code>describe(include='all')</code>

B Use <code>describe()</code> with no arguments

C Use <code>describe(numeric_only=True)</code>

D Use <code>describe(exclude='all')</code>

Using include='all' includes all columns in the summary regardless of data type.

Explain what the describe() function does in pandas and what kind of summary statistics it provides for numerical data.

Think about the quick overview it gives for numbers in your data.

How can you use describe() to get summary statistics for categorical columns? What information does it show?

Consider how categorical data differs from numerical data.

Practice

1. What does the describe() function in pandas primarily provide for a DataFrame?

easy

A. A statistical summary including count, mean, std, min, and max

B. A list of all column names in the DataFrame

C. A plot of the data distribution

D. A sorted version of the DataFrame

Solution

1. Step 1: Understand the purpose of describe()

   The describe() function gives a quick statistical summary of numerical columns by default.

2. Step 2: Identify what the summary includes

   This summary includes count, mean, standard deviation, minimum, maximum, and quartiles.

3. Final Answer:

   A statistical summary including count, mean, std, min, and max -> Option A

4. Quick Check:

   describe() = statistical summary [OK]

Hint: Remember: describe() shows stats, not column names or plots [OK]

Common Mistakes:

• Thinking describe() lists column names
• Confusing describe() with plotting functions
• Assuming describe() sorts data

2. Which of the following is the correct way to get a statistical summary of a pandas DataFrame named df?

easy

A. df.summary()

B. df.describe()

C. describe(df)

D. df.stats()

Solution

1. Step 1: Recall pandas method syntax

   In pandas, methods are called on DataFrame objects using dot notation.

2. Step 2: Identify the correct method name

   The method to get a statistical summary is describe(), so df.describe() is correct.

3. Final Answer:

   df.describe() -> Option B

4. Quick Check:

   Method call on DataFrame = df.describe() [OK]

Hint: Use dot notation: df.describe() for summary [OK]

Common Mistakes:

• Using non-existent methods like summary() or stats()
• Calling describe() as a standalone function
• Missing parentheses after describe

3. Given the DataFrame:

import pandas as pd
data = {'A': [1, 2, 3, 4, 5], 'B': [5, 4, 3, 2, 1]}
df = pd.DataFrame(data)
print(df.describe())

What is the output for the mean of column A?

medium

A. 1

B. 2.5

C. 15

D. 3.0

Solution

1. Step 1: Calculate the mean of column A

   Column A values are [1, 2, 3, 4, 5]. The mean is (1+2+3+4+5)/5 = 15/5 = 3.0.

2. Step 2: Confirm output from describe()

   The describe() method prints the mean as 3.0 for column A.

3. Final Answer:

   3.0 -> Option D

4. Quick Check:

   Mean of [1,2,3,4,5] = 3.0 [OK]

Hint: Add values and divide by count for mean [OK]

Common Mistakes:

• Confusing sum with mean
• Using median instead of mean
• Misreading the output values

4. What is wrong with this code snippet to get a statistical summary of a DataFrame df?

summary = df.describe

medium

A. Missing parentheses to call the method

B. DataFrame does not have describe method

C. describe() requires arguments

D. Variable name 'summary' is invalid

Solution

1. Step 1: Check method call syntax

   The code assigns df.describe without parentheses, so it assigns the method itself, not the result.

2. Step 2: Identify the fix

   Adding parentheses df.describe() calls the method and returns the summary.

3. Final Answer:

   Missing parentheses to call the method -> Option A

4. Quick Check:

   Method call needs () to execute [OK]

Hint: Always add () to call methods like describe() [OK]

Common Mistakes:

• Forgetting parentheses after method names
• Thinking describe needs arguments
• Confusing method with variable name errors

5. You have a DataFrame df with both numeric and categorical columns. How can you use describe() to get a summary that includes statistics for all columns?

hard

A. df.describe(include='numeric')

B. df.describe(include='categorical')

C. df.describe(include='all')

D. df.describe(include=None)

Solution

1. Step 1: Understand default behavior of describe()

   By default, describe() summarizes only numeric columns.

2. Step 2: Use include='all' to summarize all columns

   Passing include='all' makes describe() include numeric and categorical columns.

3. Final Answer:

   df.describe(include='all') -> Option C

4. Quick Check:

   include='all' shows all columns summary [OK]

Hint: Use include='all' to summarize all column types [OK]

Common Mistakes:

• Using include='numeric' which is default
• Using include='categorical' which is invalid
• Passing include=None which defaults to numeric only