What is Diagnostic Assessment?

A diagnostic assessment answers one question: what does a student actually believe about a concept, and how much of that belief is wrong? It is not concerned with grading performance. It exists to surface prior knowledge, readiness, and misconceptions before they get in the way of new learning.

That makes diagnostic assessment distinct from both ends of the assessment cycle. A formative vs. summative assessment framework treats formative work as ongoing feedback during instruction and summative work as a final measure of outcomes. Diagnostic assessment sits earlier than both. Its job is to map the terrain before teaching begins, so instruction can be built on what students already know instead of what a syllabus assumes they know.

Why Diagnostic Work Exists

Universities have used diagnostic tools for decades: placement tests, concept inventories, entry surveys. Research from the University at Buffalo's Office of Curriculum, Assessment and Teaching Transformation notes that instructors routinely build course design around assumptions about incoming knowledge, and those assumptions are frequently wrong. Determining what students actually know, and what they misunderstand, before teaching a topic prevents flawed prior knowledge from compounding as new material is layered on top of it. Read more from UB CATT.

The stakes are higher than they sound. A widely cited finding from concept inventory research on biological misconceptions shows that these gaps often go undetected by the exams institutions already run, because most summative testing is built to measure retention and factual recall rather than conceptual mastery assessment. A student can pass a final exam while still holding onto the exact misunderstanding a diagnostic tool would have flagged in week one. See the concept inventory research.

The Shift Toward Precision

Diagnostic assessment used to mean a single score on a general readiness scale. A newer strand of research, cognitive diagnostic assessment, moves away from that. Instead of asking how much a student knows overall, it asks which specific misconceptions a student holds, so remediation can target the actual error rather than the general topic. A 2025 systematic review covering a decade of research on this approach found that its core value lies in mapping conceptual gaps with enough specificity to guide targeted correction, not just a readiness label. Read the cognitive diagnostic assessment review.

Higher education institutions are applying the same logic to entire transition points. Research on mathematics readiness across universities in the US, Ireland, Australia, and South Africa found that diagnostic testing works best when it identifies specific gaps early and connects them directly to targeted support, rather than functioning as a single placement gate. See the mathematics transitions study.

Diagnostic Work Doesn't Have to End Before Instruction

The convention is that diagnostic assessment happens before teaching starts. But the same logic, identifying precise misconceptions rather than a general score, is just as valuable after instruction. The real question an instructor needs answered is not just what a student believed coming in, but which of those beliefs are still wrong once teaching is supposed to be done.

This is the function Atlas performs inside Axiom Flow. Before a session begins, Atlas analyzes the uploaded learning material and generates a configurable number of misconceptions (5, 10, 15, or 20), then maps one exam question to each. That mapping is diagnostic work in the strictest sense: it defines exactly which misunderstandings the session is designed to detect, before a single word of teaching happens.

Atlas takes no part in the teaching itself. Sam, the AI student, starts each session holding the misconceptions Atlas generated and has no independent way to check what's true. The student's explanations are Sam's only source of correction. This phase is unscored, so it functions as genuine, low-stakes practice rather than a graded event.

Once teaching ends, Sam answers Atlas's exam questions using only what he was taught, with no outside reasoning to fall back on. Atlas then produces a score along with a gap report: which misconceptions were resolved, and which ones the student never actually corrected. That gap report is the diagnostic payoff, arriving after instruction instead of before it, built from misconception-based evaluation rather than a single aggregate number.

Most tools marketed to instructors as a formative assessment platform monitor whether a student recalls the right answer during practice. Axiom Flow goes further by keeping a record of the exact misconceptions Atlas mapped at the start and checking, after teaching, which of them survived. The result combines a formative phase, the unscored teaching, with a summative phase, the scored exam, so the diagnostic detail isn't lost between them the way it typically is when pre-tests and final exams are built separately.

For instructional designers, that gap report answers the question diagnostic assessment was always meant to answer: not just where a student started, but exactly where the misunderstanding still lives once the lesson is finished.