Spotting the Odd: A Beginner’s Quiz on Outlier Detection and Treatment — Questions & Answers

This quiz contains 5 questions. Below is a complete reference of all questions, answer choices, and correct answers. You can use this section to review after taking the interactive quiz above.

  1. Question 1: Identifying Outliers Using Standard Deviations

    Which technique identifies outliers in a dataset as points lying more than 3 standard deviations from the mean, such as a test score of 100 when the class average is 60 and standard deviation is 10?

    • A. Z-Score Method
    • B. Median Imputation
    • C. Cluster Sampling
    • D. K-means Outlier
    • E. Linear Regression
    Show correct answer

    Correct answer: A. Z-Score Method

  2. Question 2: Treatment by Capping (Winsorization)

    If high-income values above a certain threshold are replaced with the value at the 95th percentile, which outlier treatment technique is being used?

    • A. Transformation
    • B. Winsorization
    • C. Bootstrapping
    • D. Mean Centering
    • E. Pruning
    Show correct answer

    Correct answer: B. Winsorization

  3. Question 3: Visual Detection with Boxplots

    When a data analyst uses a boxplot to visually detect outliers, which characteristic typically reveals an outlier, such as a dot or asterisk beyond the 'whiskers'?

    • A. Tall boxes
    • B. Short whiskers
    • C. Points outside the whiskers
    • D. Colored bars
    • E. Shaded quartiles
    Show correct answer

    Correct answer: C. Points outside the whiskers

  4. Question 4: Using the IQR Rule

    Suppose a value in a dataset is below Q1 – 1.5×IQR or above Q3 + 1.5×IQR; which rule is being applied to flag outliers?

    • A. Range Rule
    • B. Variance Test
    • C. Interquartile Range (IQR) Rule
    • D. Correlation Check
    • E. Gini Coefficient
    Show correct answer

    Correct answer: C. Interquartile Range (IQR) Rule

  5. Question 5: Imputing Outliers with Median Values

    For a dataset containing an unusually high sensor reading due to an error, which treatment replaces this outlier by using the median value of the data?

    • A. Median Imputation
    • B. Mean Division
    • C. Model Fitting
    • D. K-Nearest Repair
    • E. Interpolation
    Show correct answer

    Correct answer: A. Median Imputation