DA Data Cleaning and Preparation 1 — Questions and Answers
Question 1: What is the most common method for handling missing numeric values in a dataset?
- Delete the entire column
- Replace with the column mean, median, or mode (Correct answer)
- Replace with zero always
- Flag the dataset as unusable
Correct answer: Replace with the column mean, median, or mode
Imputing missing values with the mean, median, or mode preserves dataset size and is a standard starting point for handling missingness.
Question 2: What is data normalization?
- Removing duplicate records
- Scaling numeric values to a common range such as 0 to 1 (Correct answer)
- Sorting data alphabetically
- Converting strings to uppercase
Correct answer: Scaling numeric values to a common range such as 0 to 1
Normalization rescales features to a standard range so that no single feature dominates due to its magnitude.
Question 3: Which of the following best describes an outlier?
- A data point that is exactly equal to the mean
- A value significantly different from the rest of the dataset (Correct answer)
- A missing value in the dataset
- A duplicated record
Correct answer: A value significantly different from the rest of the dataset
An outlier is a data point that lies far outside the typical range of values and can skew analysis results.
Question 4: What does data deduplication mean?
- Adding new records to a dataset
- Identifying and removing duplicate records (Correct answer)
- Splitting a dataset into training and test sets
- Converting data types
Correct answer: Identifying and removing duplicate records
Deduplication removes redundant rows that represent the same entity, preventing double-counting in analysis.
Question 5: What is a data type mismatch issue?
- When two datasets have different numbers of rows
- When a column stores values in an incorrect format for its intended data type (Correct answer)
- When column names differ between tables
- When a dataset has no primary key
Correct answer: When a column stores values in an incorrect format for its intended data type
A data type mismatch occurs when values are stored in the wrong format, such as dates stored as strings, causing calculation errors.
Question 6: What is one-hot encoding used for in data preparation?
- Normalizing numeric columns
- Converting categorical variables into binary numeric columns (Correct answer)
- Filling missing values
- Reducing dataset dimensions
Correct answer: Converting categorical variables into binary numeric columns
One-hot encoding transforms each category level into a separate binary column (0 or 1), enabling algorithms that require numeric input to process categorical data.
DA Data Cleaning and Preparation 1
What is the most common method for handling missing numeric values in a dataset?