What Is the Difference Between Percentage and Percentile?

A percentage is a measure of quantity, helpful for comparing amounts against a total. For example, a company aiming to improve diversity in hiring may look at statistics to find that 25% of current employees are women, then recommend hiring more female employees

Percentile is a measure of rank, comparing how many alternatives an entry on the list beats. In hiring, a company looking for outstanding workers may focus on the 90th percentile for experience, which means applicants who have more experience than 90% of all applicants.

Percentile is most helpful when dealing with larger quantities of data or results whose answers are not intuitive. It’s not helpful to use this as a comparison if you only have four or five things to look at, but if you have one or two hundred choices, sorting by percentile is an effective way of finding the best options.

For example, the SAT test for college admissions has a score out of 1600, but it’s not immediately obvious how good a result of 1300 is compared to other people taking the same test. It varies slightly from year-to-year based on actual results, but that score is around the 90th percentile.

Why Is Percentile a More Accurate Measure of Variability Than Percentage?

For HR, a percentage is rarely helpful for addressing the company’s needs. To stick with the theme of hiring, it may be nice to know that 40% of candidates meet every must-have qualification, but this doesn’t help choose between them.

In contrast, percentile can help a team understand which options are better than others. It’s possible to quantify every element of a candidate’s application and use math to create a ranking of your options. Viewed holistically, this is a much better metric for understanding the variability of options.

Remember, the choice that seems like a sure winner may drop down the list once you consider other factors. This is why it’s necessary to look at a wider spectrum of choices and decide how to prioritize the relevant elements.

How Can You Use Percentile To Understand the Distribution of Data?

There are many ways to understand data by using percentiles. The most common use for HR is using percentiles to rank subjects and narrow down a list of choices. This is a fast and effective way to get rid of the worst choices, allowing staff to focus on the best potential options.

It’s worth noting that percentile is not a final decision-maker for businesses. It can help you find the top option mathematically, but some elements are difficult to quantify or are too subjective to assign ratings for. In these cases, percentile can’t sort between them because you have no way to rank them.

That’s why businesses should set a specific number of options to keep with percentiles, deciding between them manually. The exact number of options to retain depends on several factors, such as how much time you have and the number of options available.

For most companies, keeping 5% to 10% of the options is effective, but some companies can go under 1%.

What Is the Difference Between Percentile and Decile?

Percentile and decile are fundamentally similar concepts, differing in how much range they cover. Percentiles evaluate on a scale of one in a hundred, while decile evaluates on a scale of one in ten. Thus, the 90th percentile and the 9th decile mean the same thing.

Percentile is more helpful for most businesses in most situations, so it sees far more use than decile. If you have so few options that decile is a better choice, you might not have enough options that it’s worth sorting things this way to start with.

How Can You Use Percentile To Compare Two Data Sets?

There are several ways to calculate percentile ranks, but for business needs, figuring out what a result is greater than is usually the most helpful. Don’t worry if it’s been a while since you did math because this equation is much easier to use than it may look at first.

We’ll begin by calculating the rank for the percentile, which equals p(n+1). P stands for percentile here, and it’s the rank we’re trying to find expressed as a decimal out of 1. So the 50th percentile is 0.5, and the 80th percentile is 0.8.

N is the sample size or the number of different entries we’re ranking. Remember, the percentile is a method for calculating how an option compares to others, so we need to know how many others there are.

Let’s say that we’re going to calculate the 80th percentile for a group of 300 options. That means our equation is 0.8(300+1). In math, we multiply the number on the outside of the parentheses by the number from our result, so it’s 0.8*301, or 240.8. We’ll round up here to get a whole number, so the 80th percentile means an option is better than 241 out of 300 options.

We can repeat this equation for a different data set to find the relative rankings between them.

The trick here is figuring out how to rank and value each of the options at your disposal. If you’re looking at test scores, this is easy. If you’re trying to determine the relative values of experience based on the length and complexity of projects at other companies, there’s a lot more nuance to shift through.

Turning Data Into Numbers

There are many ways to transform real-world data into numbers, but you don’t need to study all of them.

For the purpose of generating percentiles, the best method is making a composite result from multiple factors. Here’s how it works in practice.

Let’s say that your company is sifting through five hundred applicants for a position. These applicants will have varying histories and attributes, and the company wants to find a way to compare them to each other. To determine how they compare, you decide to rank each applicant on four factors: experience, salary needs, education, and diversity.

Next, you decide to assign values to each of these categories as a percent out of one hundred. Experience is important for your company, so you decide to assign it 50%. This means that a perfect score for a candidate in this area will count for 50% in their final composite score. You decide to apply 20% each to salary and education, and the last 10% to diversity.

Each of these sections can have different numbers. Let’s say that a hypothetical candidate is 80/100 for experience, 7/10 on salary, 3/4 on education, and 4/5 on diversity. We want to convert all of these to percentages, so the results in order are 80%, 70%, 75%, and 80%. Now we have to make the composite result.

To do this, we multiply our result by the category’s percentage of the final score. An 80% score in a category that counts for half the result means it’s a total of 40%. Seventy percent of 20 is 14%, and seventy-five percent of 20 is 15% to the score. Finally, eighty percent of 10 is 8% for our score.

Adding all of these together, our hypothetical candidate has a final score of 77% across all categories. Once we have results from other candidates, we can use percentiles to see how many are above them and how many are below.

This is how you can use the difference between percentage and percentile to quantify results and find the best overall candidates for your company.