Academic teaching experience showing classroom learning, student feedback and continuous course improvement.

Academic Experience: How Students Evaluate Teaching and Feedback

Share

Students are reliable witnesses to many parts of their academic experience: whether marking criteria were clear, whether feedback arrived in time to use, whether staff could be reached and whether a course was organized. They are much weaker judges of how much they learned and of which teaching methods produced the most learning. Research on student evaluations of teaching (SETs) finds weak relationships with measured learning in the stronger studies, and evidence that ratings can be influenced by factors unrelated to teaching, including instructor gender and cultural background, grading leniency and course type. Student ratings are therefore best read as evidence of the experience of teaching, alongside other evidence of teaching effectiveness.

Experience of teaching and teaching effectiveness are different questions

Two questions are often merged in university reporting. The first is how students experience teaching, feedback, assessment and academic support. The second is how effective that teaching is, usually judged by what students learn. Student surveys are well suited to the first question, because students are the only people who experience every part of a course. The second question needs other evidence, such as assessment results that account for prior attainment, peer review of teaching and analysis of course design. Measured teaching quality and learning outcomes belong to the Quality of Education area; this article is about the student side and about how far student reports can be stretched.

National surveys show what the student side usually covers. The questionnaire for the National Student Survey (NSS) 2027, published by the Office for Students, which manages the survey on behalf of the UK funding and regulatory bodies, groups its core questions into themes including teaching on my course, learning opportunities, assessment and feedback, academic support, organisation and management, learning resources and student voice. Its assessment and feedback questions ask how clear the marking criteria were, how fair marking and assessment have been, how well assessments allowed students to demonstrate what they learned, how often feedback was received on time and how often feedback helps students improve their work. Its academic support questions ask how easy it was to contact teaching staff and how well staff supported learning. Each is a question about experience, and each can be answered honestly by a student who cannot judge whether the course taught them well.

What the research says about ratings and learning

Feeling of learning versus actual learning

The clearest experimental evidence comes from Deslauriers, McCarty, Miller and colleagues (2019, Proceedings of the National Academy of Sciences). In large introductory physics courses at one US research university, 149 students were randomly assigned to receive the same content through either active instruction or polished lectures by experienced, highly rated instructors, with the conditions swapped for a second topic. Students in active classes scored almost half a standard deviation (0.46) higher on a test of learning, but reported more than half a standard deviation (0.56) lower feeling of learning. The authors attribute part of this negative relationship to the greater cognitive effort that active learning requires, and they caution that evaluating instruction through students' perceptions of learning could favor less effective passive methods. The study covers one subject at one institution, but its design is unusually strong for this field.

Multisection studies and the size of the relationship

The traditional way to test whether ratings reflect learning is the multisection study: several sections of the same course share a common final exam, and section-average ratings are correlated with section-average exam performance. Stroebe (2020, Basic and Applied Social Psychology) reviews this literature. He reports that an influential early meta-analysis found an average correlation of 0.43 between overall instructor ratings and section-average grades, but that later reanalyses showed small studies with few sections tended to produce very high correlations. In his summary of a later meta-analysis of 97 multisection studies, the estimated correlation fell to 0.12 after controlling for small-study effects, and to minus 0.06 when only studies that controlled for prior knowledge or ability were included. Those figures come from Stroebe's review of other work, which this article could not open directly.

Spooren, Brockx and Mortelmans (2013, Review of Educational Research) reviewed SET research published in peer-reviewed journals since 2000 through the lens of validity, covering the dimensionality debate, bias and questionnaire design. Their abstract concludes that research had not yet provided clear answers to several critical questions about validity and that many stakeholders are not convinced of the usefulness of SETs for either formative or summative purposes. On their reading, the question of validity remained open rather than settled.

Even a valid rating can be unfair when used alone

Esarey and Valdes (2020, Assessment & Evaluation in Higher Education) set the bias debate aside and asked what happens under the most favorable conditions supported by the literature: ratings that are moderately correlated with teaching quality, highly reliable and free of discrimination. Their simulation found that a large difference in scores still fails to reliably identify the better of two teachers, and that more than a quarter of faculty rated at or below the 20th percentile are above the median in instructional quality. Using multiple imperfect measures, including but not limited to SETs, produced fairer results than ratings alone. This is a modeling study, but it shows that the problem is not only bias; it is also the imprecision of a single indicator.

Bias: what the evidence supports and where it is contested

Fan, Shepherd, Slavich and colleagues (2019, PLOS ONE) analyzed 523,703 individual student surveys collected over 2010 to 2016 across five faculties of a large public university in Australia, covering 2,392 courses and 3,123 teachers. They found potential bias against women and against teachers from non-English-speaking backgrounds, with effects that varied by faculty and were strongest in some faculties, notably Science and Business. They suggest that bias may decrease where minority groups are better represented in the workforce. The authors state that the data show association, not cause: if ratings measured teaching quality, the only alternative to bias would be that women, or teachers with non-English-speaking backgrounds, were generally weaker teachers across a large population, for which they found no credible support.

Kreitzer and Sweet-Cushman (2022, Journal of Academic Ethics) reviewed more than 100 articles on bias in SETs. According to their abstract, women and other marginalized groups face significant biases in standard evaluations, but the effect of gender is conditional on other factors, and SETs have low or no correlation with learning. Heffernan (2022, Assessment & Evaluation in Higher Education) synthesized the literature more forcefully, arguing that evaluations are influenced by sexist, racist and homophobic prejudice and by discipline, and that abusive comments in open-text responses are directed mostly at women and marginalized academics. Both were read from their abstracts only. The careful reading is that bias is documented across many settings, but its size and direction depend on discipline, course type, student population and the interaction of instructor and student characteristics; it is not a constant that can be subtracted.

A second strand concerns grades. Stroebe (2020) reviews research indicating that students reward lenient grading and easier courses with more positive ratings and choose courses that promise good grades. He argues that using SETs for hiring and promotion therefore gives instructors an incentive to grade leniently and reduce workload. The empirical associations are well documented in his review; the claim that SET use causes grade inflation is his argument from that evidence.

Feedback and assessment: what students value and what helps them learn

Feedback is one area where student experience and measured effectiveness can be connected, because the research distinguishes types of feedback. Wisniewski, Zierer and Hattie (2019, Frontiers in Psychology) meta-analyzed 435 studies (994 effect sizes, more than 61,000 participants) across education levels from kindergarten to university. The average effect of feedback on learning was medium (d = 0.48), but heterogeneity was large. High-information feedback, which tells learners about the task, the process and how to improve, had the largest average effect (d = 0.99 across 42 effect sizes), while simple reinforcement or punishment had a small one (d = 0.24 across 39). The study is not specific to universities, but the pattern is relevant to any feedback item on a student survey: a student can receive a lot of feedback that carries little usable information.

Winstone, Nash, Parker and Rowntree (2017, Educational Psychologist) systematically reviewed 195 outputs published between 1985 and early 2014 on how learners receive and use feedback. They argue that the effect of feedback depends on the learner, the sender, the message and the context, and they propose a taxonomy of four recipience processes that interventions target: self-appraisal, assessment literacy, goal-setting and self-regulation, and engagement and motivation. They also note that while quality assurance surveys might suggest that more feedback is the answer, some higher education students report feeling overwhelmed by large amounts of it.

Little, Dawson, Boud and Tai (2024, Assessment & Evaluation in Higher Education) scoped interventions designed to improve students' feedback literacy. Only 16 studies met their criteria, 11 of them published in 2021; 15 involved undergraduates and nine measured feedback literacy with self-report questionnaires. They found evidence that interventions improved some elements, such as attitudes and students' confidence and agency in the feedback process, but called for better study and intervention design. The implication for measurement is that a survey item such as "feedback helps me improve" reflects both the feedback given and the student's capacity to use it.

Advising and access to faculty

Contact with teaching staff is the part of academic experience that students can judge most directly. The NSS asks how easy it was to contact teaching staff when needed. The National Survey of Student Engagement (NSSE), a research instrument run by Indiana University's Center for Postsecondary Research, takes a different approach: its Student-Faculty Interaction indicator asks how often students talked about career plans with a faculty member, worked with faculty on activities other than coursework, discussed course topics outside class and discussed their academic performance. Its Effective Teaching Practices indicator asks whether instructors clearly explained course goals and requirements, taught in an organized way, used examples to explain difficult points, gave feedback on drafts and gave prompt and detailed feedback on completed work. Frequency questions of this kind describe what happened rather than how satisfied the student was, which makes them easier to compare with records such as office-hour availability or advising appointments.

No peer-reviewed study of academic advising outcomes was reviewed in full for this article, so no claim is made here about which advising models improve outcomes. How student participation and contact can be measured more broadly is discussed in the companion article on measuring student engagement beyond attendance.

An editorial framework for reading academic-experience evidence

The table below is an editorial framework proposed by this article; it is not the UNIRANKS methodology and does not describe how UNIRANKS evaluates institutions. For each part of the academic experience it shows what a typical student survey question can establish, what it cannot, and what independent evidence could be read alongside it.

Student reports and complementary evidence on academic experience (editorial framework)
AreaTypical student questionWhat students can report reliablyWhat the answer cannot showComplementary evidence
TeachingHow good are staff at explaining things?Clarity, organization, approachability as experiencedHow much students learned; whether the method is effectiveLearning gains on common assessments with prior attainment controlled; peer observation; course design review
Assessment criteria and fairnessHow clear were the marking criteria? How fair has marking been?Whether criteria were communicated and understoodWhether marking was accurate and consistentModeration and second-marking records; external examiner or reviewer reports; appeals data
Feedback timelinessHow often was feedback received on time?Experienced turnaround against expectationsWhether a policy deadline was met across all coursesRecorded return dates against published deadlines
Feedback usefulnessHow often does feedback help you improve?Perceived usefulnessInformation content of feedback; the student's capacity to use itSamples of feedback reviewed for task and process information; feedback literacy activities in the curriculum
Access to staffHow easy was it to contact teaching staff?Experienced accessibilityActual availability or response timesPublished contact hours, advising appointments offered and taken, response-time records
Advising and mentoringHow often did you discuss your performance or career plans with a faculty member?Frequency of contactQuality of advice or its effect on outcomesAdvising caseloads, follow-up records, outcomes for advised students interpreted with caution

Using student evaluations responsibly: our interpretation

The research supports a middle position. Student ratings are not meaningless: they are the main source of evidence on clarity, organization, timeliness, fairness as experienced and access to staff, and students' views on these matters deserve weight. But the stronger studies do not support treating ratings as a measure of how much students learn, and the experimental evidence shows that perceived learning can move in the opposite direction from measured learning. Bias findings add a second reason for caution, and the simulation by Esarey and Valdes adds a third: even a valid, unbiased rating is too imprecise to rank individual teachers on its own.

Two practical consequences follow. First, items that ask about specific, observable experiences (Was feedback returned on time? Could you reach staff?) are more defensible than global items (How good was the teaching?), and they can be checked against records. Second, open-text comments need moderation for abusive content before they are shared with staff, given the evidence summarized by Heffernan. For interpreting survey response rates and percentages, see the companion article on what student satisfaction survey results really mean.

Practical steps

  1. For academic leaders: do not use student ratings as the sole or decisive evidence in hiring, promotion or contract decisions. Combine them with peer review, teaching materials, course design and learning evidence, and look at patterns over several years and courses.
  2. For survey designers: prefer specific, behavior-based questions about feedback, assessment and contact over global satisfaction items, and keep "my experience of teaching" separate from "how much I learned."
  3. For quality teams: check survey results on timeliness and access against administrative records, and investigate gaps in either direction.
  4. For analysts: before comparing ratings across staff or departments, examine differences by instructor gender, background, discipline, course level, class size and grade distribution, and report uncertainty.
  5. For faculty: explain to students why active methods may feel harder, as Deslauriers and colleagues recommend, and build feedback literacy into courses so that feedback can be used.
  6. For students: when comparing programs, ask how feedback is given and how quickly, how marking is checked for consistency, and how and when you can reach teaching staff. Answers to these questions are more informative than an overall satisfaction percentage.

Limitations

This article draws on ten peer-reviewed sources. Six were read in full (Deslauriers and colleagues; Fan and colleagues; Little and colleagues; Stroebe; Winstone and colleagues; Wisniewski, Zierer and Hattie) and four from their abstracts only (Esarey and Valdes; Heffernan; Kreitzer and Sweet-Cushman; Spooren, Brockx and Mortelmans). The multisection correlations reported here are taken from Stroebe's review rather than from the original meta-analyses, which could not be opened during this review. The experimental evidence comes from one subject at one institution; the large bias study comes from one Australian university and is correlational; and the feedback meta-analysis covers all education levels, not only higher education. Most SET research comes from English-speaking countries. Academic advising was not reviewed in depth. The framework is editorial and has not been validated as an instrument.

Related reading on UNIRANKS

For related research, see the Student Experience & Satisfaction research hub.

Editorial disclosure

UNIRANKS publishes university rankings, and its methodology includes a Student Experience & Satisfaction dimension. This article reviews independent research on how students evaluate teaching, feedback, assessment and academic support. It does not describe how UNIRANKS scores institutions.

Research References

  1. Deslauriers, L., McCarty, L. S., Miller, K., et al. (2019). Measuring actual learning versus feeling of learning in response to being actively engaged in the classroom. Proceedings of the National Academy of Sciences, 116(39), 19251-19257. https://doi.org/10.1073/pnas.1821936116 (verified 22 November 2026)
  2. Esarey, J., & Valdes, N. (2020). Unbiased, reliable, and valid student evaluations can still be unfair. Assessment & Evaluation in Higher Education, 45(8), 1106-1120. https://doi.org/10.1080/02602938.2020.1724875 (verified 22 November 2026)
  3. Fan, Y., Shepherd, L. J., Slavich, E., et al. (2019). Gender and cultural bias in student evaluations: Why representation matters. PLOS ONE, 14(2), e0209749. https://doi.org/10.1371/journal.pone.0209749 (verified 22 November 2026)
  4. Heffernan, T. (2022). Sexism, racism, prejudice, and bias: a literature review and synthesis of research surrounding student evaluations of courses and teaching. Assessment & Evaluation in Higher Education, 47(1), 144-154. https://doi.org/10.1080/02602938.2021.1888075 (verified 22 November 2026)
  5. Kreitzer, R. J., & Sweet-Cushman, J. (2022). Evaluating Student Evaluations of Teaching: a Review of Measurement and Equity Bias in SETs and Recommendations for Ethical Reform. Journal of Academic Ethics, 20(1), 73-84. https://doi.org/10.1007/s10805-021-09400-w (verified 22 November 2026)
  6. Little, T., Dawson, P., Boud, D., et al. (2024). Can students' feedback literacy be improved? A scoping review of interventions. Assessment & Evaluation in Higher Education, 49(1), 39-52. https://doi.org/10.1080/02602938.2023.2177613 (verified 22 November 2026)
  7. Spooren, P., Brockx, B., & Mortelmans, D. (2013). On the Validity of Student Evaluation of Teaching. Review of Educational Research, 83(4), 598-642. https://doi.org/10.3102/0034654313496870 (verified 22 November 2026)
  8. Stroebe, W. (2020). Student Evaluations of Teaching Encourages Poor Teaching and Contributes to Grade Inflation: A Theoretical and Empirical Analysis. Basic and Applied Social Psychology, 42(4), 276-294. https://doi.org/10.1080/01973533.2020.1756817 (verified 22 November 2026)
  9. Winstone, N. E., Nash, R. A., Parker, M., et al. (2017). Supporting Learners' Agentic Engagement With Feedback: A Systematic Review and a Taxonomy of Recipience Processes. Educational Psychologist, 52(1), 17-37. https://doi.org/10.1080/00461520.2016.1207538 (verified 22 November 2026)
  10. Wisniewski, B., Zierer, K., & Hattie, J. (2019). The Power of Feedback Revisited: A Meta-Analysis of Educational Feedback Research. Frontiers in Psychology, 10, 3087. https://doi.org/10.3389/fpsyg.2019.03087 (verified 22 November 2026)

Sources

Share this article