Employer feedback can tell universities how a particular group of employers perceives graduates and how employers behave when they screen applications. It cannot, on its own, measure the quality of the education those graduates received. The answer depends on which kind of evidence is meant: opinion surveys of employers in general, ratings by supervisors of specific graduates, hiring experiments that record callbacks, or direct assessments of graduates' skills. Each reaches a different population, carries different biases and answers a different question. Read together and with their limits stated, they are useful; a single "employers are satisfied" percentage is not.
Four kinds of evidence that get called "employer feedback"
- General employer opinion surveys ask organizations about skills, shortages and their views of graduates as a group. Respondents may or may not have hired a recent graduate.
- Supervisor ratings of named graduates ask the direct supervisor of a specific employed graduate to rate that person's skills and preparation.
- Hiring evidence records what employers do rather than what they say, most rigorously through correspondence (audit) studies that send matched fictitious applications to real vacancies and compare callback rates.
- Direct skill assessments are not employer feedback at all, but they test the skills that employers say they value, which makes them an essential cross-check.
The sections below examine what each can establish, using published survey methodology and peer-reviewed research.
Who actually answers an employer survey?
The first question to ask of any employer finding is who the employers were. Two large public surveys show why.
Most employers have not recently hired a graduate
The UK Employer Skills Survey 2022, commissioned by the Department for Education and conducted by telephone with 72,918 employers across the UK (overall response rate 53%), asked establishments whether they had recruited anyone leaving education in the previous two to three years. In England, Wales and Northern Ireland, 14% of establishments had recruited a university leaver. An opinion about "graduates" from the other employers in such a sample is therefore not based on recent direct experience. Employer surveys that report general views without separating these groups risk mixing informed and uninformed judgments.
Supervisor surveys are more direct but highly selective
Australia's Employer Satisfaction Survey (ESS), part of the Quality Indicators for Learning and Teaching (QILT) program funded by the Australian Government Department of Education and administered by the Social Research Centre, is designed to address that problem. Employed graduates who respond to the national Graduate Outcomes Survey are asked to provide contact details for their direct supervisor, who is then surveyed. The 2025 methodological report states that many other employer surveys "report the perceptions of executives who may have had little or no direct experience with recent graduates."
The same report shows the cost of this design. Of 92,663 employed graduate respondents in the 2025 population frame, 5.2% passed on valid supervisor contact details, down from 8.6% in 2024. The survey achieved 2,399 completed supervisor responses, a response rate of 49.3% of in-scope supervisors (a definition the report notes is narrower than industry standards). Completed responses therefore equal about 2.6% of employed graduates in the frame (2,399 of 92,663, our calculation). The report also identifies field differences: supervisors of Education graduates and, to a lesser degree, Engineering graduates were overrepresented, while supervisors of Health, Creative arts and Society and culture graduates were underrepresented. The report includes comparison tables to identify the extent to which the achieved sample departs from a representative survey of employers of all recent graduates. In our interpretation, because graduates decide whether to involve their supervisor, the referral step itself may filter which employment relationships are rated.
What "satisfaction" means in practice
In the 2025 ESS national report, Overall Satisfaction was 84.2%, with Technical Skills rated highest at 94.0%. Overall Satisfaction is defined as the share of supervisors who said they were very likely or likely to consider hiring another graduate from the same course and institution if they had a relevant vacancy. That is a measure of willingness to rehire, reported by a self-selected group of supervisors. It is informative, but it is not a direct test of what the graduate learned at university.
What employers say they want, and why it varies
Employer opinion research consistently lists broad capabilities. Tushar and Sooraksa (2023, Heliyon) conducted a semi-systematic review covering 30 years of English-language research articles and government reports, considering 25 studies. They identified 87 unique skills after removing duplicates, with problem-solving, communication, teamwork, adaptability and willingness to learn among the most commonly reported over time, and they report a mismatch between employer expectations and graduates' skills. The review also describes evidence that required skills differ from industry to industry and recommends research on industry-specific needs.
Who is asked changes the answer. Succi and Canovi (2020, Studies in Higher Education) compared students' and employers' views of soft skills in several European countries. According to the abstract, 86% of respondents reported an increased emphasis on soft skills over the previous five to ten years, companies considered soft skills more important than students or graduates did, and the two groups prioritized the 20 listed skills differently.
Stated skill lists also leave out how hiring decisions are actually made. In a qualitative study of 42 manufacturing firms in the US state of Wisconsin, Hora (2020, Higher Education) found, according to the abstract, that 74% of employers hired for cultural fit, matching applicants' dispositions and competencies to the personalities of existing staff and to industry-specific norms within departments. The author highlights the potential for discriminatory practices in this process. When fit to a team's existing culture drives hiring, an employer's satisfaction with new hires partly reflects that match, not only the graduate's education.
Are employer complaints about skills a reliable signal?
Reports of a graduate "skills gap" often rest on employer surveys. Cappelli (2015, ILR Review) examined such claims for the US labor force. According to the abstract, very little evidence was consistent with complaints of a skills shortage, a wide range of evidence suggested the complaints were not warranted, and overeducation remained a persistent condition of the US labor force. The article considers several possible explanations for employer complaints. Its scope is one country and a period before 2015, so it does not settle whether skills gaps exist elsewhere or now. What it does establish is that employer complaints and actual skill shortages need to be tested separately, with labor-market data such as wages, vacancies and qualifications held.
What hiring experiments can reveal
Correspondence studies measure behavior instead of opinion, and random assignment of resume attributes allows causal conclusions about how employers respond to those attributes.
Deming, Yuchtman, Abulafi and colleagues (2016, American Economic Review) randomly assigned the sector and selectivity of the issuing institution on fictitious resumes sent to real business and health vacancies on a large online job board. According to the abstract, a business bachelor's degree from a for-profit online institution was 22% less likely to receive a callback than one from a nonselective public institution. In health jobs, for-profit credentials received fewer callbacks unless the job required an external quality signal such as an occupational license.
Lennon (2021, ILR Review) sent 1,891 applications to real job openings using 100 fictitious applicant profiles designed to resemble recent graduates of established universities. According to the abstract, profiles randomly assigned a traditional in-person degree received nearly twice as many callbacks as those indicating an online degree.
Within the credentials, applicant profiles and job markets studied, these experiments provide credible evidence of how employers respond to selected signals on a resume. They do not show whether those reactions are accurate. A lower callback rate for one credential type may reflect real differences in graduates, employer stereotypes, or both, and the design cannot tell which. Callbacks are also only the first step of hiring; the studies cited here report callbacks, not wages or job performance. In the Deming study, the callback disadvantage associated with for-profit credentials was not observed in health jobs requiring a valid occupational license. This suggests that externally verified qualifications may influence how employers interpret educational credentials. The use of audit and quasi-experimental designs to study internships is discussed separately in Do University Internships Improve Graduate Career Outcomes?
Direct skill assessments as a cross-check
If employers value generic skills such as critical thinking and written communication, those skills can be assessed directly. The OECD report Does Higher Education Teach Students to Think Critically? (2022) presents results from the CLA+ International Initiative, which tested critical thinking, written communication and related skills in higher education institutions in six countries (the United States, the United Kingdom, Italy, Mexico, Finland and Chile) between 2016 and 2021, using an instrument developed by the Council for Aid to Education. The report's abstract states that a tertiary degree is "a poor indicator of skills level." Its executive summary describes the challenges of student recruitment and motivation in a pilot in the United Kingdom, and it notes that the earlier OECD AHELO feasibility study did not proceed to a main study because the 2015 proposal did not attract sufficient support.
Direct assessment avoids employer sampling problems, but it has its own: participation is often voluntary and low stakes, the institutions taking part are self-selected, and a test of generic skills says little about discipline-specific or professional competence. It complements employer evidence; it does not replace it.
Comparing the evidence
| Evidence type | What it measures | Main sources of bias | What it cannot show |
|---|---|---|---|
| General employer opinion survey | Perceived skills and shortages among responding organizations | Respondents without recent graduate hires; senior respondents far from day-to-day work; nonresponse; industry mix | How a specific institution's graduates perform |
| Supervisor rating of a named graduate | Direct supervisor's view of one employed graduate, including willingness to rehire | Graduates choose whether to refer; only employed graduates included; field over and underrepresentation | How unemployed graduates or non-referred graduates compare |
| Correspondence (audit) study | Employer callback response to randomized credential signals | Limited to callbacks, selected occupations and job boards, one labor market at a time | Whether employer responses are accurate, or effects on wages and performance |
| Direct skill assessment | Tested generic skills of participating students | Voluntary, low-stakes participation; self-selected institutions; test motivation | Discipline-specific skill or performance at work |
| Hiring and outcome records | Who is hired, where and on what terms | Selection into universities and fields; local labor markets | Whether education caused the outcome |
How to read a claim that "employers rate our graduates highly"
The following checklist is an editorial framework proposed by this article; it is not the UNIRANKS methodology and does not describe how UNIRANKS evaluates institutions.
- Who responded? Supervisors of specific graduates, recruiters, or senior executives answering in general terms? How were they selected?
- What share of the eligible population is represented? Look for the number contacted, the number responding and the denominator behind any percentage.
- Which graduates are covered? Only employed graduates, only those who agreed to refer a supervisor, or all graduates?
- What was asked? A rehire intention, a skill rating, or general satisfaction? The exact wording matters.
- Is the field mix reported? Results that pool disciplines can reflect the field composition of respondents rather than education quality.
- Is there an independent check? Behavioral hiring evidence, direct assessment or longer-term outcomes that confirm or contradict the opinion data.
Our interpretation
Read together, the sources suggest three conclusions. First, employer opinion is a perception from whichever employers answer, and the design of large public surveys shows how small and selective that group can be. Second, what employers say they want (broad, transferable skills) and how they act (screening by credential signals and cultural fit) are related but not identical, so both deserve measurement. Third, none of these sources isolates the contribution of a university's teaching. Employer views of graduates also reflect who enrolled, which fields they studied, the local labor market and the employer's own hiring preferences. This is why employer feedback is best treated as one strand in a wider evidence base on graduate preparation, alongside reach and use of career support (discussed in How Can Universities Measure Career Services Effectiveness?), graduate outcomes and direct evidence of learning.
Practical steps
For universities collecting employer feedback
- Ask employers first whether they have hired or supervised a recent graduate of the institution, and report experienced and inexperienced respondents separately.
- Publish the sampling frame, number invited, number responding and field composition with every result.
- Prefer specific, behavior-based questions (for example, about tasks the graduate performs) over general satisfaction items.
- Report results by field and occupation rather than one institutional figure.
- Use employer comments to inform curriculum review, but test proposed changes against student learning evidence rather than adopting them on opinion alone.
For researchers and quality agencies
- Combine supervisor ratings with outcome data for non-referred and non-employed graduates to estimate referral bias.
- Extend correspondence studies beyond callbacks where feasible, and to a wider range of countries and occupations.
- Link direct skill assessments to later employment to test whether the skills employers name actually predict outcomes.
Limitations
This article draws on six peer-reviewed sources. One was read in full (Tushar and Sooraksa) and five from their abstracts only (Deming and colleagues; Lennon; Succi and Canovi; Hora; Cappelli). Claims from those five stay within their abstracts; sample sizes not stated in an abstract are not reported. Two of the six were published before 2020 (Deming and colleagues in 2016 and Cappelli in 2015) and are included because they remain central to the methods discussed. The official survey evidence comes from two countries, the UK and Australia, whose survey designs differ, so their figures are not comparable with each other. The two audit studies concern online and for-profit credentials and do not test other institutional attributes. The OECD assessment involved a limited set of participating institutions in six countries. The checklist and comparison table are editorial tools and have not been validated as measurement instruments.
Related reading on UNIRANKS
For research on graduate outcomes, internships, career support and progression, see the Employability & Career Success research hub. Evidence on teaching and learning outcomes is collected in the Quality of Education research hub.
Editorial disclosure
UNIRANKS publishes university rankings, and its methodology includes an Employability & Career Success dimension. This article reviews independent research on employer feedback about university graduates. It does not describe how UNIRANKS scores institutions.
Research References
- Cappelli, P. H. (2015). Skill Gaps, Skill Shortages, and Skill Mismatches. ILR Review, 68(2), 251-290. https://doi.org/10.1177/0019793914564961 (verified 22 November 2026)
- Deming, D. J., Yuchtman, N., Abulafi, A., et al. (2016). The Value of Postsecondary Credentials in the Labor Market: An Experimental Study. American Economic Review, 106(3), 778-806. https://doi.org/10.1257/aer.20141757 (verified 22 November 2026)
- Hora, M. T. (2020). Hiring as cultural gatekeeping into occupational communities: implications for higher education and student employability. Higher Education, 79(2), 307-324. https://doi.org/10.1007/s10734-019-00411-6 (verified 22 November 2026)
- Lennon, C. (2021). How Do Online Degrees Affect Labor Market Prospects? Evidence from a Correspondence Audit Study. ILR Review, 74(4), 920-947. https://doi.org/10.1177/0019793919899943 (verified 22 November 2026)
- Succi, C., & Canovi, M. (2020). Soft skills to enhance graduate employability: comparing students and employers' perceptions. Studies in Higher Education, 45(9), 1834-1847. https://doi.org/10.1080/03075079.2019.1585420 (verified 22 November 2026)
- Tushar, H., & Sooraksa, N. (2023). Global employability skills in the 21st century workplace: A semi-systematic literature review. Heliyon, 9(11), e21023. https://doi.org/10.1016/j.heliyon.2023.e21023 (verified 22 November 2026)
Sources
- 2025 Employer Satisfaction Survey Methodological Report. Quality Indicators for Learning and Teaching, Australian Government Department of Education (Social Research Centre). https://www.qilt.edu.au/docs/default-source/default-document-library/2025-ess-methodological-report.pdf (accessed 22 November 2026)
- 2025 Employer Satisfaction Survey National Report. Quality Indicators for Learning and Teaching, Australian Government Department of Education. https://www.qilt.edu.au/docs/default-source/default-document-library/2025-ess-national-report.pdf (accessed 22 November 2026)
- Employer Skills Survey 2022: Research report. Department for Education (IFF Research). https://assets.publishing.service.gov.uk/media/65855506fc07f3000d8d46bd/Employer_skills_survey_2022_research_report.pdf (accessed 22 November 2026)
- Does Higher Education Teach Students to Think Critically? OECD (2022). https://www.oecd.org/en/publications/does-higher-education-teach-students-to-think-critically_cc9fa6aa-en.html (accessed 22 November 2026)
