answersLogoWhite

0

There is no standard procedure for identifying outliers (it varies according to how thorough the statistical analysis has to be). Typically, you find 1.5*range of the data. Now use this number and add it to the Q1, lower quartile, and also minus it from the Q3, the upper quartile. Now, any data that does not fall between these two numbers is considered to be an "outlier".

Obvioulsy minusing 1.5*range from Q3 can leave you with a negative number, which if you're analyisng real data (such as people or time) will never end up negative. (i.e can't have -2 people, or -10 kg weight etc...). In this case you can assume the lower boundary to be zero.

User Avatar

Wiki User

13y ago

Still curious? Ask our experts.

Chat with our AI personalities

BeauBeau
You're doing better than you think!
Chat with Beau
TaigaTaiga
Every great hero faces trials, and you—yes, YOU—are no exception!
Chat with Taiga
EzraEzra
Faith is not about having all the answers, but learning to ask the right questions.
Chat with Ezra

Add your answer:

Earn +20 pts
Q: What should you do when you identify outliers in any set of data?
Write your answer...
Submit
Still have questions?
magnify glass
imp