The Employer Reputation Index is a key metric of the QS World University Ranking(s)® carrying a weighting of 10% in the World ranking and different weights in other rankings we produce.
The Employer Reputation component is unique amongst current international evaluations in taking into consideration the important component of employability. The majority of undergraduate students leave university in search of employment after their first degree, making the reputation of their university amongst employers a crucial consideration.
Source of Respondents
The results are based on the responses to a survey distributed worldwide academics from a number of different sources:
- Previous Respondents
- Submitted contact lists from institutions (see Survey Nominations Procedure)
- Sign-ups on our sign-up facility (see Survey Nominations Procedure)
- Survey partners
The survey is sent to many thousands of global employers each year. It has largely followed the same principles since inception, with some variation depending on career market themes of interest over time. At the beginning of the survey, academics state their field specialism and their regional familiarity. We ask the following questions of each respondent:
- Their name
- Their company
- Their location
- The nature and size of their organization
Top Domestic Institutions
- Employers are asked to nominate up to 10 institutions that they rate as being the best for producing graduates.
Top International Institutions
- Employers are asked to nominate up to thirty international institutions from their country/territory of knowledge that they rate as being the best for producing graduates.
Each year, the survey has additional questions that seek to answer a particular theme or need from the market, or to provide deeper insight into the nominations we receive. Past themes have included online learning, sustainability, and emerging skills.
Data cleaning and validity checks
Once the survey has been collated, a variety of checks and balances are performed to ensure the responses are valid, useable and complete.
We first remove the following:
- Submitted surveys with no nominations
- Survey was completed too quickly as to be trustworthy
- Responses submitted by any QS office
- Responses submitted by students or academics (ineligible for the Employer Survey)
We then seek to ensure that the nominations provided are being done in good faith. This anomaly testing includes, but is not limited to, removing responses where:
- The same email address was used multiple times
- Multiple email addresses are associated to the same full name & nationality
- The same IP addresses have been used in both surveys
Where we see an above average number of repetitive nominations from a single IP address, we will seek to identify the respondent and investigate.
Finally, we sort and match string entries and map them to the correct variables in our analyses.
Step by Step Analysis
Once the responses have all been processed, we apply the following procedures for all of the employer nominations (with no breakdown by five broad faculty areas as is the case for Academic Reputation).
- Devise weightings based on the regional familiarity of respondents. Respondents are able to relate to more than one region.
- Devise weightings based on the location with which respondents consider themselves familiar. Here we look at how many well enough recognized institutions in the location per response originated from it. Therefore nominations originating from mature markets are weighted higher. Locations with a low participation rate are exempted from this to avoid small number effects.
- Now, having regional and location weightings ready, we derive a weighted count of international nominations for each institution. Here, we use a 5-year aggregation of nominations, where the earlier two years count for 25% and 50%, and the most recent three years at full 100% weight.
- Derive a weighted count of domestic nominations for each institution (excluding self-nominations). This is adjusted against the number of institutions from that country with a certain level of international nominations and the total response from that country. Countries with more recognized institutions naturally face more competition in terms of gaining nominations, and this is designed to reflect and reward this.
- Apply a straight scaling to each of these to achieve a score out of 100.
- Combine the two scores with equal (50:50) weightings.
- Square root the result to draw in the outliers.
- Scale the rooted score to present a score out of 100.