DSE Exploratory Data Analysis Techniques 4 — Questions and Answers
Question 1: Which method is commonly used to detect outliers using the IQR?
- Flag values beyond mean ± 1 standard deviation
- Flag values below Q1 − 1.5×IQR or above Q3 + 1.5×IQR (Correct answer)
- Flag values in the top and bottom 5% by rank
- Flag values with z-scores beyond ±1
Correct answer: Flag values below Q1 − 1.5×IQR or above Q3 + 1.5×IQR
The Tukey fence method flags as outliers any values more than 1.5×IQR below Q1 or above Q3, and this threshold is also used by box plots.
Question 2: In a pair plot (scatter plot matrix), what does the diagonal typically display?
- Scatter plots of each variable against itself
- A histogram or KDE of each variable's distribution (Correct answer)
- Correlation coefficients between variables
- Bar charts of categorical variables
Correct answer: A histogram or KDE of each variable's distribution
The diagonal of a pair plot shows the univariate distribution (histogram or KDE) of each variable since a scatter plot of a variable against itself is uninformative.
Question 3: What is the main advantage of using a KDE (Kernel Density Estimate) over a histogram for EDA?
- It is computationally faster to compute
- It produces a smooth continuous density curve that is less sensitive to bin width choices (Correct answer)
- It shows exact counts of observations in each interval
- It requires no assumption about the underlying distribution
Correct answer: It produces a smooth continuous density curve that is less sensitive to bin width choices
KDE creates a smooth density estimate that avoids the arbitrary bin-width problem of histograms, giving a clearer view of the distribution shape.
Question 4: You have a dataset with 10 features and want to explore relationships between all pairs. How many scatter plots would a full scatter plot matrix contain?
- 10
- 45
- 100 (Correct answer)
- 90
Correct answer: 100
A scatter plot matrix for n variables contains n² cells total, so 10×10 = 100 cells (10 diagonal + 90 off-diagonal scatter plots).
Question 5: Which visualization technique is specifically designed to compare distributions across multiple categorical groups simultaneously?
- Line chart
- Grouped box plot (Correct answer)
- Scatter plot
- Pie chart
Correct answer: Grouped box plot
A grouped box plot displays side-by-side box plots for each category, enabling direct comparison of median, spread, and outliers across groups.
Question 6: What does a mosaic plot (Marimekko chart) visualize?
- The distribution of a single continuous variable
- The joint distribution of two or more categorical variables (Correct answer)
- Time-series trends for multiple groups
- Correlation between two continuous variables
Correct answer: The joint distribution of two or more categorical variables
A mosaic plot tiles rectangles whose widths and heights are proportional to the marginal and conditional frequencies of two categorical variables.
Question 7: When you see a strong linear pattern on a scatter plot, the Pearson correlation coefficient (r) is close to which value?
- 0
- −1 or +1 (Correct answer)
- 0.5
- 2
Correct answer: −1 or +1
Pearson r ranges from −1 to +1; values near −1 or +1 indicate a strong linear relationship, while values near 0 indicate no linear correlation.
DSE Exploratory Data Analysis Techniques 4
Which method is commonly used to detect outliers using the IQR?