Graded Quiz: Data Wrangling :Data Analysis with Python (Applied Data Science Specialization) Answers 2025

1. Replace missing value for continuous attribute

• ❌ Use an educated guess

• ❌ Use the mean square error

• ❌ Use the difference between min and max

• ✅ Use the average of the other values in the column

Explanation: For continuous numeric data, the standard method is replacing missing values with the mean.

2. First step before deciding bin values

• ❌ Divide average by standard deviation

• ✅ Visualize the distribution with a histogram

• ❌ Convert object types

• ❌ Use IQR

Explanation: You must first understand the data’s distribution before deciding how to bin it.

3. Best data type for inconsistent city names (“N.Y.”, “Ny”, “New York”)

• ✅ object

• ❌ float

• ❌ DataFrame

• ❌ int

Explanation: Text data should be stored as object type in Pandas.

4. Primary purpose of normalization

• ❌ Make features identical

• ❌ Remove outliers

• ✅ Ensure features have similar ranges for fair comparison

• ❌ Remove missing values

Explanation: Normalization rescales features so no feature dominates because of its scale.

5. Convert categorical values for ML

• ❌ Converts numerical to categorical

• ✅ Turns categorical values into numerical values

• ❌ Divide values into bins

• ❌ Change data type

Explanation: Encoding converts categories (e.g., “red”, “blue”) into numeric form for ML algorithms.

6. First step in data preparation

• ✅ Cleaning missing or inconsistent values

• ❌ Normalizing values

• ❌ Running models

• ❌ Encoding categorical variables

Explanation: Data cleaning always comes before transformation or modeling.

7. Prepare “fuel type” column (“gas”, “diesel”)

• ❌ cut()

• ✅ get_dummies()

• ❌ dropna()

• ❌ astype()

Explanation: get_dummies() performs one-hot encoding for categorical variables.

8. Convert “N/A” to NaN

• ✅ replace()

• ❌ astype()

• ❌ dropna()

• ❌ fillna()

Explanation: replace("N/A", np.nan) converts placeholder strings to actual missing values.

🧾 Summary Table