Why Faculty Distrust Course Evaluations — and How to Win Them Back

Bottom line up front: Course-evaluation programmes are usually designed around students — how to raise response rates, protect anonymity, and show students their feedback mattered. But the people the data is supposed to move are faculty, and the evidence shows many of them distrust evaluation feedback, experience it as a judgement on their identity, and either ignore it or change their teaching for the wrong reasons. An evaluation system that the teaching staff do not believe in cannot improve teaching, no matter how high its response rate. Winning faculty trust is not a soft add-on — it is the mechanism by which evaluation produces value.

The half of the system everyone forgets

"Closing the loop" is almost always framed toward students: tell them what changed because of their feedback so they keep responding. That matters. But there is a second loop that gets far less attention — the one between the data and the academic who is meant to act on it. If faculty do not trust the instrument, do not engage with the report, or react defensively to it, the entire apparatus of collecting student voice produces nothing. And the research suggests this failure is common.

What the evidence says about faculty trust

Faculty scepticism is not irrational; it tracks the genuine weaknesses of legacy instruments. Faculty are acutely aware that small swings in scores can affect contract renewal, and that the numbers conflate teaching quality with things outside their control — and they are right to be, given the bias and noise literature. As one synthesis of the trust problem puts it, what faculty distrust most is "the isolated and anonymous variety of feedback that lacks a relationship": a decontextualised number from an unknown subset of students, delivered without dialogue.

The emotional dimension is just as real and just as under-managed. A 2022 article in the medical-education literature, "When students' words hurt", documents that faculty often interpret evaluation feedback "as a judgment not just on their teaching ability but on their personal and professional identity," and that critical comments — even constructively worded ones — can produce "disappointment, hurt, and shame" that actively hinders the reflection and innovation evaluation is supposed to spark. Defensive reception is not a character flaw; it is the predictable result of delivering anonymous criticism without support.

Crucially, distrust distorts behaviour even when faculty do act. Flodén's (2017) study of teachers at the University of Gothenburg found that while staff generally saw student feedback as impactful, those who received negative feedback experienced more negative emotion and were "more likely to introduce unjustified changes to their teaching in order to please students." That is the worst outcome the bias literature warns about — feedback driving instructors toward leniency and crowd-pleasing rather than learning — and it is rooted in how the feedback is experienced, not just what it says.

The vicious cycle this creates

These dynamics compound. Online evaluations already suffer lower response rates than in-class ones — on the order of 29% online versus 70% in class in some analyses — which gives faculty a legitimate representativeness objection. Low trust then lowers faculty effort to encourage participation, which lowers response rates further, which deepens the distrust. Meanwhile a thin, anonymous, numbers-only report gives faculty nothing constructive to act on, so they disengage, so students see no change, so students stop responding. The student-facing and faculty-facing failures feed each other.

But isn't some faculty distrust just defensiveness?

The strongest counterargument is that faculty resistance is partly self-serving — nobody enjoys criticism, and "I distrust the instrument" can be a convenient shield against legitimate negative feedback. There is truth here, and it should not be waved away: some distrust is motivated reasoning, and a well-run system must still hold teaching to account. But two things follow. First, the existence of motivated distrust does not make the instrument's real weaknesses imaginary — noise, bias, low response rates, and decontextualised reporting are documented facts, and faculty pointing at them are often correct. Second, and more practically, defensiveness is itself a design problem to solve, not a reason to dismiss the audience. If the way feedback is delivered reliably triggers shame and self-justification, the predictable result is non-engagement, regardless of how fair the underlying data is. A system that is technically valid but psychologically unusable still fails. The goal is not to indulge defensiveness but to design feedback faculty can actually receive and act on.

What wins faculty back

The evidence points to a consistent set of moves. Deliver rich, contextual feedback — specific, thematically organised, actionable — rather than a bare average, so faculty have something to do. Pair summative scores with formative, mid-cycle feedback the instructor owns privately, which research consistently finds more useful for actual improvement and far less threatening. Triangulate so no single noisy number carries decisive weight, which directly answers the fairness objection. Be transparent about the instrument's limits, which paradoxically builds trust. And support reception — frame feedback as developmental, not just evaluative.

How Koji fits

Koji for Education is built to repair the faculty side of the loop, not only the student side. Its AI-moderated conversational interviews and automatic thematic analysis turn raw comments into structured, specific themes an instructor can act on — replacing the bare, identity-threatening average with a constructive, organised account of what students actually experienced. By making formative, mid-cycle collection as easy as end-of-term summative evaluation, Koji gives faculty private, developmental feedback they own, which the evidence ties to genuine teaching improvement and lower defensiveness. Because Koji surfaces distributions, themes, and quality scores rather than a single ranked mean, it supports the triangulated, uncertainty-aware reporting that answers faculty's fairness objection head-on. And its closing-the-loop action tracking works in both directions — not only showing students what changed, but giving departments a record of how teaching responded. The honest framing: Koji makes feedback more usable and trustworthy and reduces the decontextualised, identity-threatening quality of legacy reports — it cannot make every instructor welcome criticism, and it does not pretend to. (The same engine underpins stakeholder and employee research on the main Koji platform, where trust in the instrument is just as decisive.)

What a faculty-first evaluation cycle looks like in practice

Reorienting toward faculty is less about a new instrument than a new choreography. A faculty-first cycle starts before the summative window with a low-stakes, mid-term check the instructor controls and reads first, so the first encounter with student feedback in a term is private, developmental, and within their power to act on. The summative report then arrives with context — themes rather than a lone mean, the response rate stated plainly, and like-for-like benchmarks rather than the cross-disciplinary league tables faculty rightly distrust. Heads of department are equipped to discuss the feedback as evidence to be interpreted, not a score to be defended. And the loop closes visibly on the faculty side too: a short record of what the instructor changed, which both demonstrates the system works and protects staff acting in good faith on noisy data.

This choreography is also what makes the student-facing loop credible. Students keep responding when they see change; change happens when faculty trust and act on the data; faculty trust the data when it arrives as developmental evidence rather than anonymous judgement. The two loops are not competing priorities — they are the same loop viewed from both ends, and neglecting the faculty end is why so many otherwise well-run programmes stall.

Universities spend enormous effort persuading students to fill in evaluations and far less ensuring faculty believe, engage with, and act on what comes back. That asymmetry is why so many evaluation programmes generate compliance without improvement. The institutions that get this right tend to share one habit: they treat faculty as the primary users of evaluation data rather than its subjects, and they measure the system's success by whether teaching changed, not by how many forms were returned. Win the faculty back, and the data finally does what it was always supposed to: change teaching.

See how Koji turns student feedback into something faculty actually trust and use — explore Koji for Education.