Data Science with Python
MCQs
1. In the "Data Hierarchy of Needs" pyramid, which layer must be solid before
you can attempt AI or Deep Learning?
A. The Middle (Analytics/Metrics)
B. The Top (Predictive Modeling)
C. The Bottom (Data Collection/Infrastructure)
D. The Side (Data Visualization)
2. Using the "Car Analogy," how does Data Science differ from Data
Analytics?
A. Data Science is the engine; Analytics is the wheels.
B. Data Science is the rearview mirror (what happened); Analytics is the GPS
(what will happen).
C. Data Science is the GPS (what will happen); Analytics is the rearview mirror
(what happened).
D. There is no difference; they are synonyms.
3. According to the "Data Kitchen" analogy, which library acts as the "Prep
Station" where you chop, peel, and clean data?
A. NumPy
B. Pandas
C. Scikit-Learn
D. Matplotlib
4. Why did Amazon’s AI hiring tool fail (The "Black Mirror" Moment)?
A. The math in the algorithm was incorrect.
B. It was trained on historical data that contained human bias against women.
C. It used Unstructured data instead of Structured data.
D. The dataset was too small to make a prediction.
5. Which of the following is NOT one of the three main goals of EDA?
A. Sanity Check (Is the data broken?)
B. Assumption Testing (Is it normally distributed?)
C. Model Training (Fitting the final algorithm)
D. Feature Selection (Which columns matter?)
6. Why is the Median often preferred over the Mean when analyzing salaries?
A. The Median is easier to calculate.
B. The Mean is too sensitive to outliers (e.g., if Bill Gates walks into a bar).
C. The Mean only works for categorical data.
D. The Median provides a wider range of values.
7. You want to visualize the distribution of a single numerical variable (e.g.,
Student Ages) to see if it follows a Bell Curve. Which chart should you use?
A. Bar Chart
B. Scatter Plot
C. Histogram
D. Pie Chart
8. In the "Chart Chooser" framework, which plot is recommended for
comparing a Numerical variable against a Categorical variable (e.g., Salary vs.
Gender) to spot outliers?
A. Scatter Plot
B. Box Plot (Box-and-Whisker)
C. Line Chart
D. Heatmap
9. What is the fundamental difference between Non-Statistical and Statistical
Analysis?
A. Non-Statistical is probabilistic; Statistical is deterministic.
B. Non-Statistical deals with samples; Statistical deals with populations.
C. Non-Statistical is deterministic (facts); Statistical is probabilistic (estimates).
D. Non-Statistical uses Python; Statistical uses Excel.
10. In the "Soup Analogy," what does the "Spoonful" represent?
A. The Population
B. The Sample
C. The Outlier
D. The Variable
11. A survey about internet usage conducted only on Twitter is an example
of:
A. Random Sampling
B. Sampling Bias
C. Confounding Variables
D. Validity Testing
12. Which branch of statistics is described as the "Crystal Ball" that allows
you to generalize from a sample to a population?
A. Descriptive Statistics
B. Inferential Statistics
C. Predictive Analytics
D. Non-Statistical Analysis
Unit IV: Advanced Statistical Analysis
13. In a perfectly Normal Distribution (The Bell Curve), which of the following
is true?
A. Mean > Median
B. Mean < Mode
C. Mean = Median = Mode
D. There is no Mode.
14. In Hypothesis Testing (The Courtroom Analogy), if the P-Value is very low
(< 0.05), what is the verdict?
A. The defendant is innocent (Fail to Reject Null).
B. The evidence is inconclusive.
C. The Null Hypothesis must go (Reject Null / Significant Result).
D. The Alternative Hypothesis is false.
15. You calculate a correlation coefficient ($r$) of 0.8 between Ice Cream
Sales and Drowning Rates. What is the most likely explanation?
A. Eating ice cream causes drowning.
B. Drowning causes people to buy ice cream.
C. It is a random coincidence ($r$ is too low to matter).
D. There is a Confounding Variable (Summer Heat) causing both.
16. Which measure of dispersion brings the variability back into the original
units of the data (e.g., "dollars" instead of "dollars squared")?
A. Variance
B. Range
C. Standard Deviation
D. Interquartile Range (IQR)
Unit V: Numerical & Mathematical Computing
17. Why is NumPy faster than standard Python lists?
A. It uses C-language optimization and Vectorization.
B. It uses larger memory blocks.
C. It skips the validation step.
D. It only works with integers, not floats.
18. If you slice a NumPy array (e.g., b = a[0:5]), what does NumPy create by
default?
A. A Deep Copy (independent object)
B. A View (reference to original data)
C. A Tuple
D. A JSON object
19. Which NumPy attribute would you check to see if an array is a Vector
(1D), Matrix (2D), or Tensor (3D)?
A. .shape
B. .dtype
C. .ndim
D. .size
20. If NumPy is the "Brick and Mortar" (Foundation), what is SciPy?
A. The Plotting Tool.
B. The Database.
C. The Architecture (Advanced mathematical algorithms built on top).
D. The Cleaning Station.
Answer Key
1. C (Bottom: Collection/Infrastructure)
2. C (Science = GPS; Analytics = Rearview)
3. B (Pandas)
4. B (Historical bias in training data)
5. C (Model Training happens after EDA)
6. B (Mean is sensitive to outliers)
7. C (Histogram)
8. B (Box Plot)
9. C (Non-stat = Deterministic; Stat = Probabilistic)
10.B (The Sample)
11.B (Sampling Bias)
12.B (Inferential Statistics)
13.C (Symmetry means they are equal)
14.C (Low P = Reject Null)
15.D (Confounding Variable)
16.C (Standard Deviation)
17.A (Vectorization)
18.B (A View - changing it changes the original!)
19.C (.ndim)
20.C (Architecture/Advanced tools)