What is Skewness?

S8-SA1-0133

Grade Level:

Class 6

AI/ML, Data Science, Research, Journalism, Law, any domain requiring critical thinking

Definition

What is it?

Skewness tells us if data is lopsided or symmetrical. It shows if most values are clustered on one side, with a 'tail' extending to the other side. Think of it like a seesaw – is it balanced or heavier on one end?

Simple Example

Quick Example

Imagine marks of students in a math test. If most students scored low marks (like 20-30 out of 100) and only a few scored very high (like 90-100), the data would be 'skewed' towards the lower marks. The 'tail' would be on the higher mark side.

Worked Example

Step-by-Step

Let's look at the daily auto-rickshaw fares collected by a driver in a week: 50, 60, 55, 150, 70, 65, 50.

1. Plot these values roughly on a number line. You'll see most fares are around 50-70 rupees.

2. Notice the fare of 150 rupees. It's much higher than the others.

3. If you draw a curve over these points, it would have a peak around 50-70 rupees.

4. The curve would then stretch out towards the higher value of 150, forming a 'tail' on the right side.

5. This means the data is 'skewed to the right' or 'positively skewed' because the tail is on the right side (higher values).

ANSWER: The auto-rickshaw fare data is positively skewed.

Why It Matters

Understanding skewness helps data scientists and AI/ML engineers make better predictions by knowing how data is distributed. Journalists use it to correctly interpret survey results, and researchers use it to understand patterns in information, helping them draw accurate conclusions about important topics.

Common Mistakes

MISTAKE: Thinking skewness means the data is 'wrong' or 'bad' | CORRECTION: Skewness is just a description of the data's shape. It's not good or bad, it's just how the data naturally is.

MISTAKE: Confusing the direction of the tail with the direction of the peak | CORRECTION: Skewness is named after the direction of the 'tail' (the longer part of the curve), not where the main hump is.

MISTAKE: Assuming all data should be perfectly symmetrical | CORRECTION: Many real-world datasets are naturally skewed. For example, income data is usually positively skewed because a few people earn very high salaries.

Practice Questions

Try It Yourself

QUESTION: If most students in a class score high marks in a test, and only a few score very low, what kind of skewness would the marks data likely show? | ANSWER: Negatively skewed (tail towards lower marks)

QUESTION: A shop sells 100 items. 90 items are priced between ₹50-₹100, and 10 items are priced between ₹500-₹1000. Is the pricing data positively or negatively skewed? | ANSWER: Positively skewed (tail towards higher prices)

QUESTION: Imagine a list of waiting times (in minutes) for a popular food stall: 2, 3, 4, 5, 5, 6, 7, 8, 9, 30. If you draw a graph, where would the 'tail' be? What kind of skewness is this? | ANSWER: The tail would be on the right side (towards 30 minutes). This is positive skewness.

MCQ

Quick Quiz

What does it mean if data is 'positively skewed'?

Most data points are high, with a tail towards lower values.

Most data points are low, with a tail towards higher values.

The data is perfectly symmetrical.

The data has two peaks.

The Correct Answer Is:

Positive skewness means the 'tail' of the data distribution points to the right, towards higher values. This happens when most data points are concentrated on the lower end.

Real World Connection

In the Real World

When a company like Swiggy or Zomato analyzes delivery times, they might find the data is positively skewed. Most deliveries are quick, but a few take much longer due to traffic or bad weather. Understanding this helps them manage expectations and improve service.

Key Vocabulary

Key Terms

SYMMETRICAL: Balanced, like a mirror image on both sides | TAIL: The longer, thinner part of a data distribution | POSITIVE SKEWNESS: Tail is on the right (higher values) | NEGATIVE SKEWNESS: Tail is on the left (lower values)

What's Next

What to Learn Next

Next, you can learn about 'Measures of Central Tendency' like Mean, Median, and Mode. Skewness helps us understand why sometimes the Mean, Median, and Mode are different from each other!