Module 2 Study Material – Descriptive Analytics
(Analogy-Based)
This guide covers every item from your checklist, presented with practical analogies for each
concept. Let’s work from foundations through advanced subtopics, so you can link abstract
statistical concepts to real-world experiences, making them easier to master and recall during
exams.
1. Descriptive Analytics – Foundations
What is Descriptive Analytics?
Analogy: Think of descriptive analytics as the summary screen on a fitness tracker: it tells you
your steps, calories, and sleep averages for the week, but doesn’t tell you why they changed or
predict next week. In business, it summarizes "what happened" using data—like past sales or
website visits.
How is it Different From Predictive and Prescriptive Analytics?
Descriptive: The snapshot (like reading last month’s electricity bill).
Predictive: Like a weather forecast, using patterns to estimate what might happen next
week.
Prescriptive: Like GPS rerouting you after traffic appears, it recommends the best action for
an optimal outcome.
2. Measures of Central Tendency
Mean (Arithmetic Average)
Analogy: Imagine dividing a pizza so everyone gets the same share, regardless of who ate
what before.
Formula: Mean ( : each value, : total number of values)
Advantages/Applications: Works well with symmetric data; sensitive to outliers (i.e., if one
person grabs extra pizza, the average increases for everyone).
Median
Analogy: Line all your friends by height; the person standing in the middle is the median. For an
even number, take the average of the two middle people.
Use: Not affected by outliers; preferred for skewed data, e.g., home prices (a single
mansion won’t distort the median).
Mode
Analogy: Most popular ice cream flavor in your class—the flavor ordered by most students.
There can be more than one mode (if tied), or sometimes none.
Use: Especially useful for categorical data (e.g., finding what type of error happens most
frequently on a website).
3. Measures of Dispersion
Range
Analogy: It’s the difference between the tallest and shortest players on a basketball team—a
quick sense of spread but doesn’t explain how everyone else is distributed.
Formula: Range = Max – Min
Limitation: Highly affected by outliers (someone 7ft tall on a normally average-height team).
Variance (Population/Sample)
Analogy: It’s like measuring how far each student’s grade is from the class average, then
averaging those squared differences.
Population variance: $ \sigma^2 = \frac{\sum (X_i-\mu)^2}{n} $
Sample variance: $ s^2 = \frac{\sum (X_i-\overline{X})^2}{n-1} $
Standard Deviation
Analogy: Think of it as the "average distance" backpackers stray from the group leader on a
hike (mean). Square root of the variance.
Formula: $ SD = \sqrt{Variance} $
Use: Shows average deviation in original units; crucial for understanding spread in
distributions.
Interquartile Range (IQR)
Analogy: The "middle 50% spread": If test scores are arranged in order, the IQR covers the
range between the 25th percentile (Q1) and the 75th percentile (Q3)—like the bulk of students
sitting between the two quartile benches.
Formula: IQR = Q3 – Q1
Finding Q1 and Q3: Sort data, split in quarters.
Outliers and IQR:
Lower Bound = Q1 – 1.5 × IQR
Upper Bound = Q3 + 1.5 × IQR
Scores outside bounds are flagged as outliers (like students who score way more or less
than most).
4. Data Visualization and Exploration
Categorical Variables
Bar Chart:
Analogy: Row of jars, each filled with marbles for a different category (e.g., color of cars
sold). Bar heights = counts/amounts. Easy to compare.
Pie Chart:
Analogy: Slicing a cake according to recipe use—shows how each ingredient (category)
forms part of the whole, but hard for small slices and detailed comparisons.
Numerical Variables
Histogram:
Analogy: Baskets by age group at a school picnic, showing how many children fall in each
range. Bar width = bin range.
Shape tells you symmetry, skewness, and if there are "gaps" (missing baskets) or
spikes.
Box Plot:
Analogy: Like a suitcase showing where most clothes are packed (box for middle 50%),
with "whiskers" for lighter items at the ends (extremes).
Highlights quartiles, median, and outliers as “luggage sticking out.”
Relationship/Comparison
Scatter Plot:
Analogy: Imagine placing stickers on a map where each stands for house price at a location.
Patterns emerge—more stickers near water (price vs. location).
Bubble Chart:
Analogy: Like scatter plots but each sticker’s size shows a third dimension—e.g., location
(x), price (y), and units sold (bubble size).
Time Series Data
Line Chart:
Analogy: ECG in a hospital: shows change over time (heartbeats/minute, sales/day).
Reveals uptrends, downtrends, seasonality.
Area Chart:
Analogy: Filling a glass with water over time—the area shows not just top level, but the total
volume collected as the line rises.
Multivariate/Hierarchical Data
Tree Map:
Analogy: Visualizing how a company’s total budget is split between departments, then
teams within each department—like nested boxes.
Heatmap/Facet Grid:
Analogy: A seating chart colored by exam scores: darker colors = higher scores. Displays
many variables on two axes (e.g., products vs. months).
Feel free to check off each topic as you review it! If you’d like analogies or guided examples for
any calculation or plot, let me know which one to explore together.
⁂
1. [Link]
2. [Link]
3. [Link]
4. [Link]
5. [Link]
6. [Link]