2019 GLASSDOOR U.S. EMPLOYEE RETENTION STUDY

Methodology

This survey was conducted online within the U.S. by Dynata on behalf of Glassdoor among 2,430 employees between November 8 and November 24, 2019 (percentages for some questions are based on a subset, based on their responses to certain questions). With a pure probability sample of 2,430, one could say with a 95 percent probability that the overall results have a sampling error of +/- 2.0 percentage points, respectively. Sampling error for data from subsamples is higher and varies.

Informed candidates from Glassdoor have materially higher relative annual retention rates.

• Informed candidates from Glassdoor are 15% more likely to be retained than other hires¹.
• The four-year average retention rate for employees who used Glassdoor is 63%, versus 55% for those who did not.
  • • • (63%-55%)/55% = 15% higher retention
• The overall four-year average retention rate is 57%.

Retention Rate Cohort Analysis

In order to calculate retention rates, respondents were grouped in 12-month cohorts based upon when they were hired at both their current employer and/or their last employer. Retention rates are calculated as:

• The number of respondents who started during a specific time period, e.g. Dec ‘17 – Nov ‘18, and who are still at their current employer, divided by the total number of respondents who started either their current or previous job during that same time period.

Retention Rate Formula

βᵅ = started current employer during period ɑ Ɣᵅ = started last employer during period ɑ Ρᵅ = retention rate for period ɑ Ρᵅ = βᵅ/(βᵅ + Ɣᵅ)

Sample Frame

Participants were screened to meet the following criteria:

• Ages 18-65
• Is employed full time or part-time, excluding self-employment
• OR is currently self-employed but previously was a paid employee of a company/organization within the last 5 years;
• OR is not employed but was a paid employee of a company/organization within the last 5 years

The overall incidence of respondents in the US workforce is estimated to be 50%.

• A soft quota was set to ensure a balance of male and female respondents.
• Data was weighted by gender and age group to be representative of U.S. workforce using data from the U.S. Bureau of Labor Statistics.

  ¹Based on 4-year average of December 2014 through November 2018. U.S. data.