Assessment Maker: Design Diagnostic, Formative, and Summative Assessments From One Source
An assessment maker is the umbrella tool for measuring learning across the whole instructional cycle — diagnostic (what students know before you teach), formative (check while you teach), and summative (measure at the end). Unlike a single quiz or test maker, it handles mixed item types, weighting, rubrics, and result analysis, so one source of questions can serve every stage. This guide frames assessment as a framework, not just an output file, and shows how to build it from your own validated items — not AI-fabricated ones whose mis-keyed answers would make every score meaningless.
The three stages of assessment
| Stage | When | Purpose | Typical items |
|---|---|---|---|
| Diagnostic | Before teaching | Find gaps & priors | Short MCQ, true/false |
| Formative | During teaching | Adjust instruction | Quick checks, exit tickets |
| Summative | End of unit | Measure achievement | Mixed types, weighted |
The same question bank can feed all three — a diagnostic pulls the easy tier, the summative pulls the hard tier with weights. That reuse is the real payoff of an assessment maker over a one-off quiz tool.
Assessment maker vs test maker vs quiz maker
| Quiz maker | Test maker | Assessment maker | |
|---|---|---|---|
| Scope | Short check | One-off exam | Whole measurement cycle |
| Item types | Usually MCQ/T/F | Mixed, often printable | Mixed + rubric-scored |
| Weighting | Flat | Per-item points | Section weights + rubrics |
| Analysis | Score only | Score + grade | Difficulty, discrimination, re-teach |
A quiz maker answers "did they get it?" A test maker answers "what grade?" An assessment maker answers "what do I do next?" — which is why it owns the analysis step the others skip.
The fixed-column assessment sheet
Tag each item by stage/type and weight so one file serves every purpose:
| section | type | question | option1…4 | answer | weight | image |
|---|---|---|---|---|---|---|
| Diagnostic | mcq | What is the capital of France? | Paris/Lyon/Marseille/Nice | 1 | 1 | |
| Summative | tf | Water boils at 100°C at sea level. | — | True | 2 | |
| Summative | short | Explain photosynthesis in one sentence. | — | (rubric) | 5 | diagram.png |
The section column lets you pull a subset per stage; weight drives the total score; a single image URL per question covers diagram-based items. Open-ended rows are scored with a rubric you define, not auto-graded.
Building and analyzing
Map items to stages
Tag each question diagnostic / formative / summative so the same bank serves the whole unit.
Assign weights & rubrics
Harder summative items weigh more; open responses get a rubric (e.g., 0-5 on correctness + reasoning).
Deliver
Push to Canvas, Moodle, Google Forms, or print. Mixed types export cleanly via QTI 2.1 where supported.
Compute difficulty
% correct per item: >90% too easy, <30% unclear or mistaught. Flag both extremes.
Compute discrimination
Did top-scorers get it while bottom-scorers missed it? Low discrimination means the item doesn't separate ability — rewrite it.
Act on the output
The analysis tells you what to re-teach, not just who passed. Close the loop before the next stage.
Why analysis is the differentiator
A quiz gives you a number; an assessment gives you a plan. Item difficulty shows whether the question was fair; discrimination shows whether it measured anything. Without those two, you're grading blind. Attach them to every summative and the next diagnostic writes itself.
Frequently asked questions
What is an assessment maker?
An assessment maker is the umbrella tool for measuring learning across the instructional cycle — diagnostic (what students know before), formative (check during), and summative (measure at the end). Unlike a single quiz or test maker, it handles mixed item types, weighting, rubrics, and result analysis so one source of questions can serve every stage.
How is an assessment maker different from a test maker or quiz maker?
A quiz maker produces a short check; a test maker produces a one-off exam (often printable + online). An assessment maker is broader: it frames the measurement purpose (diagnostic/formative/summative), lets you weight sections and attach rubrics, and — crucially — analyzes results afterward (item difficulty, discrimination) so you improve the next round. It is the framework, not just the output file.
How do I structure an assessment in a fixed-column sheet?
Add a 'section' or 'type' column so mixed items (MCQ, true/false, short answer, matching) live in one file, plus a 'weight' column for points per item. Keep one image URL per question where a diagram is needed. The converter emits each item in the right format and sums weights for a total score.
What should I analyze after an assessment?
Two metrics matter: item difficulty (what % got it right — too high means too easy, too low means unclear or mistaught) and item discrimination (did strong students get it while weak students didn't). Pair that with a rubric-scored section for open responses. The output tells you what to re-teach, not just who passed.
Why build assessments from my own items instead of an AI generator?
An assessment is a measurement instrument — if the items are fabricated or mis-keyed, every number it produces is meaningless. Build from items you've validated so the diagnosis, the formative check, and the final grade all rest on questions you trust.
Build assessments you can act on
Stop generating questions you can't vouch for. Upload the items where every answer is already correct, structure them across diagnostic, formative, and summative stages, and turn the results into a re-teach plan — on any LMS.
Try ExamParser →