Skip to main content icon/video/no-internet
icon backReturn to Entries

Percentile and Percentile Rank

The terms percentile and percentile rank are considered by some people to mean the same thing. However, these two terms are different in conceptual meaning and should not be used interchangeably.

Percentile Rank

A percentile rank describes the standing, or position, of an earned score in comparison to a reference group. The percentile rank for that earned score indicates the percentage of scores in the reference group that are lower than the earned score. Thus, an examinee whose test performance earns him or her a percentile rank of 72 would have scored higher than 72 percent of those in the reference group who took the same test.

Percentile ranks are reported in terms of whole numbers between 1 and 99 (inclusive). No decimals are used, and percentile ranks never assume the values of 0 or 100. These extreme values are not used because the percentile rank for any earned score considers that earned score to be in the reference group. Considered in that way, it is logically impossible for an earned score to be higher than no scores or all of the scores.

It is important to note that a percentile rank indicates relative (not absolute) standing. Thus, knowing the nature of the reference group is exceedingly important if one is to properly interpret an examinee's percentile. An examinee's percentile rank might be 92 when the reference group is a nationwide group of test takers, 81 when the reference group is defined as those in the examinee's state, but only 47 if the reference group is test takers in the examinee's school.

Percentile

A percentile is a point along the score continuum that divides the reference group's distribution of earned scores into two parts such that a desired percentage of the group's scores lie below that point. For example, if we are dealing with a normal distribution of IQ scores (with a mean of 100 and a standard deviation of 15), then 75% of the IQ scores would fall below 110.1. Thus, that score value (110.1) would be the 75th percentile.

In computing percentiles, we state a percentage (such as 75% in the above example) and then examine the distribution of scores in the reference group to determine what particular value of the score continuum has that particular percentage of scores below it. This sequence of steps is different from that used in determining a percentile rank, for in the latter case, we begin with a given earned score and then determine what percentage of the reference group's scores are below that earned score. Because of this difference, a percentile can assume a value that's not the same as any earned score. For example, if the earned scores of six examinees on a 20-item quiz in which quiz scores are determined simply as “number correct” are 19, 17, 16, 14, 13, and 12, the 50th percentile would be equal to 15, a score no one earned. (The value of 15 comes from taking the arithmetic average of the two middle scores, 16 and 14.) Pretend that the two lowest-scoring examinees are not part of the group, and in this situation, the 50th percentile for the remaining four scores is 16.5, a score that not only was not earned but also was not earnable!

...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles

Read next

More like this

Sage Recommends

We found other relevant content for you on other Sage platforms.

close

Have you created a personal profile? Login or create a profile so that you can save clips, playlists and searches

close
close
close Signed In to profile You are signed in as:
close
Logged in Institution Institution