Unit 1: Data preprocessing and visualization - Practice Quiz

1 Which of the following is an example of categorical (qualitative) data?

2 Data that can only take specific separate values, such as the number of students in a class, is called:

3 A ranking of customer satisfaction as poor, average, good, excellent is an example of which data type?

4 Which scale of measurement has a true zero point?

5 Replacing a missing numeric value with the average of the available values is known as:

6 Removing all rows that contain any missing values is referred to as:

7 In many datasets, missing values are commonly represented by which of the following?

8 Which technique estimates a missing value using neighboring known data points, often in a sequence?

9 A scatter plot is primarily used to show the relationship between:

10 In a scatter plot, if points rise together from lower-left to upper-right, the relationship is:

11 Each individual point in a scatter plot represents:

12 A histogram is mainly used to display the:

13 The intervals into which data is grouped in a histogram are called:

14 How does a histogram differ from a bar chart?

15 Group plots are most useful for:

16 A grouped bar chart is typically used to compare:

17 The line inside the box of a box plot represents the:

18 In a box plot, points that fall far outside the whiskers are typically identified as:

19 The height of the box in a box plot represents the:

20 What is the main goal of dimensionality reduction?

21 A dataset records customer satisfaction as Poor, Average, Good, and Excellent. Which data type best describes this variable?

22 Temperature measured in degrees Celsius is an example of which measurement scale?

23 You encode the number of children per household as . How should this variable be classified?

24 A numerical column has a strongly right-skewed distribution with several outliers, and about 5% of values are missing. Which imputation choice is most appropriate?

25 Values are missing more often for low-income respondents who chose not to report income. This missingness mechanism is best described as:

26 A feature is missing in 78% of records and is weakly related to the target. What is usually the most sensible action?

27 A scatter plot of two variables shows points tightly clustered along a line that falls from top-left to bottom-right. What does this indicate?

28 You want to visualize the relationship between two numeric variables while also encoding a third numeric variable. Which scatter plot enhancement is most appropriate?

29 In a large scatter plot, many points overlap and hide the true density of data. Which technique best addresses this overplotting?

30 A histogram built with too many narrow bins appears jagged and noisy. What is the primary effect of increasing the bin width?

31 A histogram shows most values bunched on the left with a long tail extending to the right. How is this distribution described?

32 Which statement correctly distinguishes a histogram from a bar chart?

33 You want to compare the distribution of test scores across five different schools on a single figure. Which grouped visualization is most suitable?

34 A faceted (small-multiples) plot arranges several subplots that share the same axes. What is its main advantage?

35 When using a grouped bar chart to compare two categories across several months, what does color (hue) typically encode?

36 In a standard box plot, the length of the box itself represents which quantity?

37 Using the common rule, a point is flagged as an outlier if it lies:

38 A box plot has , median , and . What does the position of the median inside the box suggest?

39 In Principal Component Analysis (PCA), the first principal component is chosen so that it:

40 Why is it generally recommended to standardize features before applying PCA?

41 A researcher encodes customer satisfaction as where the gaps between adjacent levels are not guaranteed to be equal. A colleague computes the arithmetic mean and standard deviation of this variable. What is the most defensible critique?

42 Temperature in Celsius and temperature in Kelvin describe the same physical quantity. Which statement correctly distinguishes their measurement scales?

43 In a dataset, income is missing more often for high earners because wealthy respondents decline to answer. The missingness depends on the unobserved income value itself. This mechanism is best classified as:

44 A data scientist replaces all missing values in a feature with the column mean before computing the sample variance. What is the primary distortion introduced?

45 Compared to single imputation, why does multiple imputation with imputed datasets produce more valid standard errors?

46 Two variables have Pearson correlation , yet a scatter plot shows a clear parabolic pattern. What is the correct interpretation?

47 When plotting 500,000 points where many overlap, a scatter plot becomes a solid ink blob (overplotting). Which technique best preserves density information?

48 A histogram of the same data looks unimodal with 10 bins but bimodal with 40 bins. What does this most likely indicate?

49 Using the Freedman–Diaconis rule, bin width is . For a sample with and , what is the approximate bin width?

50 A right-skewed variable is plotted as a histogram, then re-plotted after a transform. What effect is most expected in the transformed histogram?

51 You facet (small-multiple) a scatter plot by a categorical grouping variable across 6 panels. What is the key advantage of a shared common axis range across all panels?

52 A grouped (clustered) bar chart shows category means for three groups but omits any uncertainty indicators. What is the most serious analytical limitation?

53 Using the standard rule with and , at what upper value does a point begin to be flagged as an outlier?

54 Two box plots have identical medians, quartiles, and whisker lengths, yet their underlying distributions differ dramatically (one bimodal, one unimodal). What limitation does this reveal?

55 A notched box plot is used to compare two groups. If the notches of the two boxes do not overlap, the most appropriate conclusion is:

56 Before applying PCA to features measured in wildly different units (e.g., meters and kilograms), why is standardization typically required?

57 The eigenvalues of a covariance matrix are . What proportion of total variance is explained by the first two principal components?

58 A key difference between PCA and t-SNE for visualization is that:

59 When choosing the number of components via a scree plot, the 'elbow' heuristic suggests keeping components:

60 In a scatter plot of study hours vs. exam score, adding a third variable (course difficulty) reverses the apparent positive trend within each course subgroup. This is an example of: