What is Outlier?

S3-SA3-0050

Grade Level:

Class 8

AI/ML, Data Science, Physics, Economics, Cryptography, Computer Science, Engineering

Definition

What is it?

An outlier is a data point that is significantly different from other data points in a set. It 'lies out' or stands apart from the majority of the data, appearing much larger or much smaller than the rest.

Simple Example

Quick Example

Imagine your class scores in a Maths test: 20, 22, 25, 23, 95, 21. Here, most scores are around 20-25. But 95 is much higher than all the others. So, 95 is an outlier because it's unusually high compared to the other scores.

Worked Example

Step-by-Step

Let's find the outlier in the daily temperatures (in Celsius) recorded in a city for a week: 28, 29, 30, 29, 10, 27, 28.

1. Look at all the numbers in the data set: 28, 29, 30, 29, 10, 27, 28.
---2. Observe the general range or pattern of most numbers. Most temperatures are between 27 and 30 degrees.
---3. Identify any number that falls far outside this general range. The number 10 is much smaller than all the other temperatures.
---4. Compare 10 to the other numbers. It's almost 17-20 degrees lower than the others.
---5. Conclude that 10 is the unusual data point.

Answer: The outlier in this data set is 10.

Why It Matters

Understanding outliers is crucial in many fields. Data scientists use it to spot fraud in online transactions, engineers use it to detect faulty parts in machines, and even doctors use it to find unusual readings in patient reports. It helps us make better decisions by understanding what's normal and what's unusual.

Common Mistakes

MISTAKE: Thinking every unusual number is an outlier. | CORRECTION: An outlier must be significantly different, not just slightly higher or lower. There are statistical methods (like IQR) to formally identify them, but for now, look for a very clear 'jump' or 'drop'.

MISTAKE: Ignoring outliers and including them in calculations for average. | CORRECTION: Outliers can heavily skew the average (mean). Sometimes, they should be removed or analyzed separately to get a true picture of the main data.

MISTAKE: Confusing an outlier with a maximum or minimum value. | CORRECTION: While an outlier can be the maximum or minimum, not all max/min values are outliers. For example, in 1, 2, 3, 4, 5, 5 is the maximum but not an outlier.

Practice Questions

Try It Yourself

QUESTION: Identify the outlier in the following set of numbers representing the number of rotis eaten by family members: 2, 3, 2, 1, 10, 3, 2. | ANSWER: 10

QUESTION: A small shop's daily earnings (in Rupees) for a week were: 500, 550, 600, 520, 2000, 580, 530. Which amount is most likely an outlier? | ANSWER: 2000

QUESTION: The heights of five friends in cm are: 145, 148, 150, 152, 185. If a new friend joins whose height is 149 cm, does 185 remain an outlier? Explain. | ANSWER: Yes, 185 remains an outlier. Even with the new data point, 185 is still significantly taller than all other heights (which are between 145-152 cm).

MCQ

Quick Quiz

Which of the following data sets definitely contains an outlier?

10, 11, 12, 13, 14

5, 6, 7, 8, 50

100, 101, 102, 103

1, 2, 3, 4, 5

The Correct Answer Is:

In option B, 50 is significantly larger than the other numbers (5, 6, 7, 8), making it an outlier. In other options, numbers are relatively close to each other.

Real World Connection

In the Real World

Outliers are super important in areas like fraud detection for UPI payments. If someone suddenly makes a payment of ₹50,000 when their usual payments are ₹500-₹1000, that ₹50,000 transaction is an outlier. Banks' AI systems flag such outliers to check for suspicious activity and protect your money.

Key Vocabulary

Key Terms

DATA SET: A collection of related numbers or information | DATA POINT: A single piece of information in a data set | MEAN: The average of a set of numbers | ANOMALY: Something that deviates from what is standard, normal, or expected (similar to outlier)

What's Next

What to Learn Next

Now that you know what an outlier is, you can explore how to calculate different types of averages (mean, median, mode) and how outliers affect them. This will help you understand data better and make smarter decisions with numbers!