Statistics Definitions > What is a Decile?

Watch the video or read the article below:

Deciles are similar to quartiles. But while quartiles sort data into four quarters, deciles sort data into ten equal parts: The 10th, 20th, 30th, 40th, 50th, 60th, 70th, 80th, 90th and 100th percentiles.

A **decile rank** assigns a number to a decile:

Decile Rank | Percentile |

1 | 10th |

2 | 20th |

3 | 30th |

4 | 40th |

5 | 50th |

6 | 60th |

7 | 70th |

8 | 80th |

9 | 90th |

The higher your place in the decile rankings, the higher your overall ranking. For example, if you were in the 99th percentile for a particular test, that would put you in the decile ranking of 10. A person who scored very low (say, the 5th percentile) would find themselves in a decile rank of 1.

## Why are Decile ranks used instead of percentiles of quartiles?

Decile rankings are just another way to categorize data. Which system you use is usually a judgment call. For example, if you wanted to display class rankings on a pie chart, using deciles would make more sense that percentiles. That’s because a pie chart with 10-categories would be much easier to read than a pie chart with 99 categories.

## What is a Decile used for in Real Life?

Deciles and decile ranks are used more often in real life than in the classroom. For example, Australia uses decile ranks to report drought data. Deciles 1-2 represent the lowest 20% (“much below normal”). That means droughts that are “much below normal” don’t occur more than 20% of the time.

Deciles are also commonly used for college admissions and high school rankings. For example, this chart from Roanoke College shows the high school decile rankings for the student body.

**Next**: Interdecile Range

If you prefer an online interactive environment to learn R and statistics, this free R Tutorial by Datacamp is a great way to get started. If you're are somewhat comfortable with R and are interested in going deeper into Statistics, try this Statistics with R track.

Comments are now closed for this post. Need help or want to post a correction? Please post a comment on our Facebook page and I'll do my best to help!
Dear Sir/Madam,

-I have a research that wants to find asset growth and stock return relation.

-in each year t we grouped stocks based on their asset growth in to 10 deciles then we calculate raw return of these deciles for year t+1

-then we find spread between decile 1 and decile 10

– but in main article it says that the spread is significant at 5% level and 1% and 10%

I want to know how we find these significant level and how to calculate t-statistics for this example ?

I am waiting for reply as it is important for me.(please send me to this E-mail: maziar.etemadi2015@gmail.com

Best regards,

Mazier,

You said you wanted to find the relation between asset growth and stock return relation. You would find that by calculating a correlation coefficient. A t-statistic would tell you if you had a significant difference between samples, or a difference from a mean. But I don’t think that’s what you are after? If you do want to run a t test you’ll need a clear hypothesis statement as the first step, and you’ll need a known mean.

You may want to check these two articles out:

Correlation

T Statistic

Quoted from your article:

“The higher your place in the decile rankings, the higher your overall ranking. For example, if you were in the 99th percentile for a particular test, that would put you in the decile ranking of 9 position. A person who scored very low (say, the 5th percentile) would find themselves in a decile rank of 1”

This does NOT make any sense.

As you’ve stated, a person who scored in the 5th percentile would be in the decile of rank 1. So it follows that a person who scored between the 11th and 20th percentile would be in the decile of rank 2; a person between the 21st and 30th percentile would be in the decile of rank 3; … a person between 71st and 80th percentile would be in the decile of rank 8; a person between the 81st and 90th percentile would be in the decile of rank 9.

But then you stated that a person in the 99th percentile would be in the decile of rank 9.

So does this mean that the decile of rank 9 would include those who score between the 81st and the 99th percentile? If that is the case, then the data set would no longer be divided in 10 equal parts. If that is not that case, then either the one in 99th percentile should belong to the decile of rank 10, or the one in the 5th decile should belong to the decile of rank 0.

Thanks for your correction, Melvin. You are correct, it should have been 10 (not 9). It’s fixed.