Selection ratio in hiring: formula, base rate, Taylor-Russell and the four-fifths rule

The selection ratio in hiring is the number of people hired divided by the number of applicants: selection ratio = hires ÷ applicants. Four hires from 160 applicants is a selection ratio of 0.025, or one hire for every 40 applicants. In industrial and organizational psychology it is one of three inputs, with the assessment's validity and the base rate of successful employees, that decide how much a selection test actually improves hiring, a relationship set out in Taylor and Russell's 1939 tables.

Our pages on pipeline conversion rates and the interview-to-hire ratio cover the stage-by-stage funnel. This page covers the term as psychologists and the federal Uniform Guidelines use it, and what it means for where you spend assessment effort.

How do you calculate the selection ratio?

Selection ratio = Number hired ÷ Number of applicants

Applicants per hire = 1 ÷ Selection ratio

Both numbers are legitimate; they answer different questions. The first describes demand for the job. The second is the one that matters for assessment, because it is the pool your interviews and tests actually choose from. Say which denominator you used every time you report it, and define "applicant" the same way your applicant flow log does.

How is it different from yield ratios and the selection rate?

The Uniform Guidelines on Employee Selection Procedures define selection rate as "the proportion of applicants or candidates who are hired, promoted, or otherwise selected" (29 CFR 1607.16(R), as of October 2026). Same arithmetic as the selection ratio, but calculated per race, sex or ethnic group, as the section below shows.

What is the base rate?

The base rate is the share of applicants who would turn out to be successful employees if you hired them without any assessment at all. You cannot observe it directly for applicants, so it is usually estimated from current employees: if a manager rates 60% of the people in a role as meeting the standard, the base rate for that standard is about 0.60. The definition of "successful" sets it, which is why a written standard such as the ones in time to productivity helps.

The base rate matters because assessment helps most in the middle. If nearly everyone who applies would succeed, there are few failures to screen out. If almost nobody would, there are few successes to find.

Why does a low selection ratio raise the payoff of a valid assessment?

H. C. Taylor and J. T. Russell's paper "The relationship of validity coefficients to the practical effectiveness of tests in selection: discussion and tables" (Journal of Applied Psychology, vol. 23, no. 5, 1939, pp. 565–578) showed that a test's practical value depends on three things together: its validity (how well scores correlate with job performance), the selection ratio and the base rate. Their tables give the "success ratio", the share of people hired using the test who turn out successful.

The table below shows the pattern. It is our own calculation, using the same bivariate normal model the Taylor-Russell tables assume, for a base rate of 50%. It is an illustration of the relationship, not a benchmark for any real test.

Read down a column and three things stand out:

• With no validity, selectivity buys nothing. Rejecting 95% of applicants with an assessment that does not predict performance leaves the success rate at the base rate. You have only rejected more people.
• With a valid assessment, the lower the ratio, the bigger the gain. At validity 0.30, hiring 1 in 20 lifts the success rate from 50% to about 74%; hiring 9 in 10 lifts it to about 52%.
• Validity and selectivity multiply. A better assessment adds the most when you also have applicants to spare.

What does that mean for your hiring process?

• High-ratio roles (you hire most qualified applicants, common in high-volume and hard-to-fill work): more assessment steps add little. Put effort into sourcing more applicants, a realistic job preview so people self-select, and onboarding.
• Low-ratio roles (many applicants per opening): this is where a valid structured interview or work sample pays off. Build it from a job analysis, and see types of pre-employment assessments for the options.
• Sourcing to lower the ratio only helps if what comes next is valid. Doubling applicants into an unstructured process mostly doubles screening time.
• Count the cost of each applicant. A lower ratio means more screens, interviews and assessment fees per hire. The gain in successful hires has to outweigh them.

How does the selection rate feed the four-fifths rule?

Under the Uniform Guidelines, "a selection rate for any race, sex, or ethnic group which is less than four-fifths (4/5) (or eighty percent) of the rate for the group with the highest rate will generally be regarded by the Federal enforcement agencies as evidence of adverse impact" (29 CFR 1607.4(D), as of October 2026). The same section says smaller differences can still count when they are significant in statistical and practical terms, and larger ones may not when the numbers are small.

Selectivity interacts with this. The stricter the cut-off on an assessment, the more any difference between groups' average scores can show up as a difference in selection rates, so check selection rates by group at each step that uses a cut-off, not only at the final hire. Our pages on the four-fifths rule and disparate treatment vs. disparate impact go further. This is general information, not legal advice.

How should you report it?

SELECTION REPORT: [Role], [period]
Applicants: 160   Qualified: 40   Hires: 4
Selection ratio (all applicants): 0.025 (40 applicants per hire)
Selection ratio (qualified): 0.10
Assessment steps used: phone screen, structured interview, work sample
Selection rates by group at each step: see applicant flow log [link]
Note: base rate for "meets standard at 90 days" in current team: ~0.60

Common mistakes