# Assessment Blueprint: [Assessment title]

> Assessment blueprint (table of specifications) template for an exam, test
> or set of course assessments. It maps each learning objective against the
> six cognitive process levels of the revised Bloom's taxonomy (Anderson and
> Krathwohl, 2001), then turns the agreed marks into an item map, a timing
> plan, a review checklist and a post-test item analysis. Replace every
> `[bracketed]` value and every row marked *Example*.

<!--
Guidance: How to use this template
- Build the blueprint before writing a single item. Decide the weight of each
  objective first, then the mix of cognitive levels, then the items.
- Weight objectives by their importance and by the teaching time they
  received, not by how easy they are to test.
- Keep the matrix (section 3) and the item map (section 4) in step: every
  cell total in the matrix must equal the sum of the matching items.
- The worked example is an end-of-unit exam for an introductory statistics
  unit, 60 marks in 90 minutes. Replace it with your own assessment.
- Pair this template with `education/exam-answer-key` for the marking scheme
  and `education/rubric` for performance tasks.
-->

## Document control

<!-- Guidance: Identify the assessment and who signs off the blueprint. Approve the blueprint before item writing starts, and record any later change in the revision history. -->

| Field | Value |
|---|---|
| Document ID | [ABP-2026-007] |
| Assessment title | [End-of-unit exam, Unit 3: Describing data] |
| Course / subject | [Introductory Statistics, STAT 110] |
| Assessment type | [Summative / Formative / Diagnostic / Placement] |
| Stakes | [Low / Medium / High] |
| Total marks | [60] |
| Duration | [90 minutes, including 10 minutes reading time] |
| Mode | [Paper / Online, supervised / Online, open book] |
| Owner | [@assessment-lead] |
| Reviewers | [@second-marker], [@moderator] |
| Version | [0.1] |
| Status | [Draft / In review / Approved / Administered / Analysed] |
| Classification | [Restricted until the assessment closes] |
| Last updated | YYYY-MM-DD |

### Approvals

| Role | Name | Decision | Date |
|---|---|---|---|
| Course lead | [@name] | [Approved / Changes requested] | YYYY-MM-DD |
| Moderator or second reviewer | [@name] | [Blueprint and items checked] | YYYY-MM-DD |
| Accessibility or disability services contact | [@name] | [Accommodations confirmed] | YYYY-MM-DD |

### Revision history

| Version | Date | Author | Change summary |
|---|---|---|---|
| 0.1 | YYYY-MM-DD | [@assessment-lead] | First draft of the matrix |
| 0.2 | YYYY-MM-DD | [@assessment-lead] | *Example:* Analyse marks raised from 8 to 12 after moderator review |
| 1.0 | YYYY-MM-DD | [@assessment-lead] | *Example:* Approved for item writing |

## Contents

<!-- Guidance: Keep this list in step with the headings. -->

1. [Purpose and decisions](#1-purpose-and-decisions)
2. [Learning objectives and weights](#2-learning-objectives-and-weights)
3. [Table of specifications](#3-table-of-specifications)
4. [Item map](#4-item-map)
5. [Process](#5-process)
6. [Review checklist](#6-review-checklist)
7. [Post-assessment item analysis](#7-post-assessment-item-analysis)
8. [Glossary and references](#8-glossary-and-references)
9. [Standard and licence note](#standard-and-licence-note)

## 1. Purpose and decisions

<!-- Guidance: State what decision the scores will inform. A diagnostic quiz and a high-stakes final need different blueprints. Pitfall: using one test for several purposes without saying so. -->

| Question | Answer |
|---|---|
| What decision will the scores support? | *Example:* Grade for Unit 3 (20% of the course grade) and a list of objectives to revisit in Unit 4 |
| Who takes it? | *Example:* All students enrolled in STAT 110, about 120 per term |
| What does a pass mean? | *Example:* The student can describe and compare data sets well enough to start inference in Unit 4 |
| Pass mark and how it was set | [*Example:* 30 of 60, reviewed after item analysis] |
| Reuse | [*Example:* One new form per term; at most 30% of items reused from the previous term] |

## 2. Learning objectives and weights

<!-- Guidance: Copy the objectives from the syllabus or unit plan word for word, then agree a weight for each. A useful check: the share of marks should be close to the share of teaching time. Explain any large gap. -->

| ID | Objective | Teaching time | Weight | Marks |
|---|---|---|---|---|
| LO-1 *(Example)* | Identify types of variables and levels of measurement | 1 of 10 hours | 10% | 6 |
| LO-2 *(Example)* | Compute measures of centre and spread for a data set | 2.5 of 10 hours | 25% | 15 |
| LO-3 *(Example)* | Construct and read histograms and box plots | 2 of 10 hours | 20% | 12 |
| LO-4 *(Example)* | Interpret and compare two distributions in context | 2.5 of 10 hours | 25% | 15 |
| LO-5 *(Example)* | Judge whether a numerical or graphical summary is fair, and improve it | 2 of 10 hours | 20% | 12 |
| **Total** | | **10 hours** | **100%** | **60** |

## 3. Table of specifications

<!-- Guidance: Distribute each objective's marks across the cognitive process levels the objective actually calls for. The verb in the objective is the first clue: "compute" sits mainly in Apply, "judge" in Evaluate. Leave a cell empty rather than forcing marks into it. -->

The six levels are the cognitive process dimension of the revised taxonomy:
Remember, Understand, Apply, Analyse, Evaluate and Create. The revised
taxonomy also has a knowledge dimension (factual, conceptual, procedural and
metacognitive knowledge); add it as a column in section 4 if your team
classifies items by both dimensions.

| Objective | Remember | Understand | Apply | Analyse | Evaluate | Create | Total |
|---|---|---|---|---|---|---|---|
| LO-1 *(Example)* | 4 | 2 | | | | | 6 |
| LO-2 *(Example)* | | 3 | 12 | | | | 15 |
| LO-3 *(Example)* | | 2 | 6 | 4 | | | 12 |
| LO-4 *(Example)* | | 3 | | 8 | 4 | | 15 |
| LO-5 *(Example)* | | | | | 8 | 4 | 12 |
| **Total marks** | **4** | **10** | **18** | **12** | **12** | **4** | **60** |
| **Share** | 6.7% | 16.7% | 30.0% | 20.0% | 20.0% | 6.7% | 100% |

*Example* reading of the matrix: 14 marks (23.3%) test remembering and
understanding, 18 marks (30.0%) test application, and 28 marks (46.7%) test
analysing, evaluating and creating. Agree the target profile with your team
before writing items, and record why it suits the purpose in section 1.

## 4. Item map

<!-- Guidance: List every item with its objective, level, format and marks. Check that each matrix cell equals the sum of its items. Mark time is a planning estimate; adjust it after piloting. -->

| Item | Section | Objective | Level | Format | Marks | Planned minutes |
|---|---|---|---|---|---|---|
| Q1-Q4 *(Example)* | A | LO-1 | Remember | Multiple choice, 1 mark each | 4 | 4 |
| Q5-Q6 *(Example)* | A | LO-1 | Understand | Multiple choice, 1 mark each | 2 | 2 |
| Q7-Q9 *(Example)* | A | LO-2 | Understand | Multiple choice, 1 mark each | 3 | 3 |
| Q10-Q11 *(Example)* | A | LO-3 | Understand | Multiple choice, 1 mark each | 2 | 2 |
| Q12-Q14 *(Example)* | A | LO-4 | Understand | Multiple choice, 1 mark each | 3 | 3 |
| Q15 *(Example)* | B | LO-2 | Apply | Short answer: mean, median and standard deviation | 6 | 8 |
| Q16 *(Example)* | B | LO-2 | Apply | Short answer: quartiles and outlier rule | 6 | 8 |
| Q17 *(Example)* | B | LO-3 | Apply | Construct a box plot from a summary | 6 | 8 |
| Q18 *(Example)* | B | LO-3 | Analyse | Read a histogram and describe its shape | 4 | 5 |
| Q19 *(Example)* | B | LO-4 | Analyse | Compare two distributions in context | 8 | 11 |
| Q20 *(Example)* | C | LO-4 | Evaluate | Choose and justify a measure of centre | 4 | 6 |
| Q21 *(Example)* | C | LO-5 | Evaluate | Critique a misleading published chart | 8 | 13 |
| Q22 *(Example)* | C | LO-5 | Create | Redesign the chart from Q21 | 4 | 7 |
| **Total** | | | | **22 items** | **60** | **80** |

Section totals in the *Example*: Section A has 14 marks (14 minutes),
Section B has 30 marks (40 minutes) and Section C has 16 marks (26 minutes).
With 10 minutes of reading time, the paper fits the 90-minute slot.

## 5. Process

<!-- Guidance: Show who does what between the blueprint and the results. Keep the order: blueprint, items, review, administration, analysis. -->

The diagram shows the cycle from objectives to the next blueprint. Item
analysis results feed the item bank, so each new form starts from items with
known behaviour.

```mermaid
flowchart LR
  A["Objectives and weights"] --> B["Table of specifications"]
  B --> C["Item map"]
  C --> D["Item writing"]
  D --> E["Review against checklist"]
  E -->|"Changes needed"| D
  E -->|"Approved"| F["Administer"]
  F --> G["Item analysis"]
  G --> H["Update item bank and next blueprint"]
```

## 6. Review checklist

<!-- Guidance: A second reviewer works through this list before the paper is approved. Record the outcome, not just a tick. -->

| Check | Outcome |
|---|---|
| Every item measures the objective and level the item map says it does | [Yes / Items changed: list] |
| Matrix totals equal item map totals for every cell and every objective | [Yes / No] |
| No item gives away the answer to another item | [Yes / No] |
| Wording is plain; reading load is no higher than the content needs | [Yes / No] |
| Multiple-choice distractors are plausible and there is one best answer | [Yes / No] |
| Contexts and names are varied and free of stereotypes | [Yes / No] |
| Marking scheme or rubric exists for every constructed-response item | [Yes / Link to `education/exam-answer-key`] |
| Timing tested by a colleague, with a margin for slower readers | [Minutes taken] |
| Accommodations are in place (extra time, alternative format, assistive technology) | [List agreed accommodations] |
| Digital delivery is accessible: keyboard use, screen reader labels, zoom | [Yes / Not applicable] |

## 7. Post-assessment item analysis

<!-- Guidance: After marking, check how each item behaved. The thresholds below are common rules of thumb, not fixed standards; decide yours with your team and consider the class size. Flag items for review, never delete marks without moderation. -->

For each item, compute the facility (difficulty) index as the mean mark
divided by the maximum mark, and a discrimination figure such as the
correlation between the item score and the rest of the test.

$$
p = \frac{\text{mean mark on the item}}{\text{maximum mark for the item}}
$$

*Example:* Q19 had a mean of 4.6 out of 8, so $p = 4.6 / 8 = 0.575$.

| Item | Facility p | Discrimination | Flag | Action |
|---|---|---|---|---|
| Q19 *(Example)* | 0.575 | [Value] | None | Keep |
| [Item] | [p] | [Value] | [Too easy / Too hard / Negative discrimination / Ambiguous wording] | [Keep / Revise / Retire / Moderate marks] |

Summarise what the results say about teaching as well as about the items:
objectives with low facility across several items may need reteaching in the
next unit.

## 8. Glossary and references

<!-- Guidance: Define terms your team may read differently, and list the sources you used. -->

| Term | Meaning |
|---|---|
| Blueprint / table of specifications | A plan that sets how marks are spread across objectives and cognitive levels |
| Cognitive process dimension | The six levels of the revised Bloom's taxonomy, from Remember to Create |
| Facility index | Mean mark on an item as a share of its maximum; higher means easier |
| Discrimination | How well an item separates students who did well on the whole test from those who did not |
| Form | One version of a test built from the same blueprint |

References:

- Anderson, L. W. and Krathwohl, D. R. (eds.) (2001). *A Taxonomy for
  Learning, Teaching, and Assessing: A Revision of Bloom's Taxonomy of
  Educational Objectives*. New York: Longman (now Pearson).
- AERA, APA and NCME (2014). *Standards for Educational and Psychological
  Testing*. Cited for background on content-related validity evidence; not
  reproduced. Official text: AERA, APA and NCME, purchase or licensed access:
  https://www.testingstandards.net/purchasing-information.html (a revision
  was under way as of October 2026; check the site for the current edition).

## Standard and licence note

Structure aligned with the cognitive process dimension of the revised Bloom's
taxonomy (Anderson and Krathwohl, 2001), cited by name, and with common
practice for tables of specifications in classroom assessment. No text from
the book or from the Standards for Educational and Psychological Testing is
reproduced. This template is original text written for this collection;
example objectives, marks and statistics are illustrative. Check the current
edition of any cited publication before relying on it. Template licence: MIT.