What is Skewness? | Simple Explanation for Class 12

S7-SA3-0346

What is Skewness (Introductory)?

Grade Level:

Class 12

AI/ML, Physics, Biotechnology, FinTech, EVs, Space Technology, Climate Science, Blockchain, Medicine, Engineering, Law, Economics

Definition

What is it?

Skewness tells us about the symmetry of data. It measures if a dataset is lopsided, meaning if one side of its distribution is longer or fatter than the other side. A symmetrical dataset has zero skewness, while an asymmetrical one shows positive or negative skewness.

Simple Example

Quick Example

Imagine the daily sales of 'samosas' at your local shop. If most days they sell around 100 samosas, but some days they sell very few (like 20) and only a few days they sell a lot (like 200), the sales data might be skewed. If many days have high sales and only a few days have very low sales, it's skewed differently.

Worked Example

Step-by-Step

Let's find the type of skewness for a small dataset: 10, 20, 30, 40, 100

1. Calculate the Mean: (10 + 20 + 30 + 40 + 100) / 5 = 200 / 5 = 40
---2. Find the Median (the middle value when sorted): The sorted data is 10, 20, 30, 40, 100. The middle value is 30.
---3. Find the Mode (the most frequent value): All values appear once, so there is no clear mode, or we can say it's multimodal. For simplicity in introductory skewness, we often look at Mean vs Median.
---4. Compare Mean and Median: Mean (40) > Median (30).
---5. Conclusion: Since the Mean is greater than the Median, the data is positively skewed (skewed to the right). This means the 'tail' of the data extends more towards higher values.

Why It Matters

Understanding skewness helps scientists in AI/ML build better models by knowing how data is distributed. In FinTech, it helps analyze stock prices to understand risk. Engineers use it to study material strength, ensuring structures are safe and reliable.

Common Mistakes

MISTAKE: Thinking that a dataset with a long tail on the left is positively skewed. | CORRECTION: If the tail of the distribution extends more towards the left (lower values), it is negatively skewed. Think of the tail 'pulling' the mean to that side.

MISTAKE: Confusing skewness with kurtosis. | CORRECTION: Skewness measures the asymmetry of the distribution, while kurtosis measures the 'tailedness' or 'peakedness' of the distribution. They describe different aspects of data shape.

MISTAKE: Believing that if the mean and median are exactly equal, the data must be perfectly symmetrical. | CORRECTION: While equal mean and median often indicate symmetry, it's not always perfect. A perfectly symmetrical distribution also requires the mode to be equal to the mean and median.

Practice Questions

Try It Yourself

QUESTION: If the average height of students in a class is 150 cm, and the middle height (median) is 155 cm, what type of skewness would this data likely have? | ANSWER: Negatively skewed (Mean < Median)

QUESTION: A test had scores: 5, 10, 10, 15, 20, 25, 100. Calculate the Mean and Median. Is the data positively or negatively skewed? | ANSWER: Mean = 26.43 (approx), Median = 15. Since Mean > Median, it's positively skewed.

QUESTION: Describe a real-life scenario where you might expect to see negatively skewed data. Explain why. | ANSWER: Income distribution in a very rich neighbourhood. Most people would have very high incomes, but a few might have significantly lower incomes, pulling the mean down relative to the median.

MCQ

Quick Quiz

Which of the following statements is true for a positively skewed distribution?

The tail is on the left side.

Mean < Median < Mode

Mean > Median

The data is perfectly symmetrical.

The Correct Answer Is:

In a positively skewed distribution, the tail extends towards the right (higher values), pulling the mean in that direction, making the Mean greater than the Median.

Real World Connection

In the Real World

In cricket, analysts use skewness to understand batsman scores. If a batsman often scores very low but occasionally hits a huge century, their score distribution might be positively skewed. This helps coaches and teams strategize, like how often a player is likely to hit big.

Key Vocabulary

Key Terms

SYMMETRICAL: When both sides of a distribution are mirror images of each other. | ASYMMETRICAL: Not symmetrical; one side is different from the other. | POSITIVE SKEWNESS: The tail of the data extends towards higher values (right side). | NEGATIVE SKEWNESS: The tail of the data extends towards lower values (left side). | MEAN: The average of all values in a dataset.

What's Next

What to Learn Next

Next, you can learn about 'Kurtosis', which further describes the shape of data distributions by looking at its peaks and tails. Understanding both skewness and kurtosis gives you a complete picture of your data's shape.