> For the complete documentation index, see [llms.txt](https://docs.lamsfoundation.org/lams/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.lamsfoundation.org/lams/question-bank/essay.md).

# Essay

## Create an Essay or open text question 

Use an essay question when students need to construct, organise and communicate a response in their own words. Essay questions can assess explanation, argument, evaluation, synthesis, reflection, problem-solving and—when you enable an appropriate code style—written program code.

Choose this format when the quality of the response cannot be judged from a selected or very short answer. Do not use an essay simply to make an assessment appear more demanding. The prompt, marking criteria and expected evidence must align with the learning outcome.

Essay responses require human judgement. Plan the marking criteria, workload and moderation process before you deploy the question.

### Before you create the question

Start with the learning outcome and identify what students must demonstrate. Decide whether you need to assess:

* disciplinary knowledge;
* analysis or evaluation;
* the construction of an argument;
* application to a case or authentic situation;
* reflection on evidence or experience;
* written communication;
* a problem-solving process; or
* code design and implementation.

Draft the marking criteria alongside the question. If you cannot describe what distinguishes a strong response from a weak one, revise the task before adding it to the Question Bank.

Decide what resources and technologies students may use, including generative AI. State the permitted, restricted or required use clearly and explain any acknowledgement requirements. Do not rely on AI-detection tools as the assessment design.

### Create the question

1. Enter a short, distinctive **Question's title**.

   Use the title to identify and manage the item in the Question Bank. Include information that distinguishes it from similar items, such as the topic and intended cognitive demand. Do not put information here that students need in order to answer the question.
2. Enter the prompt in **Question's description**.

   State what students must do, the context, the expected evidence and any relevant constraints. Use an action verb that matches the learning outcome, such as **analyse**, **compare**, **evaluate**, **justify**, **design** or **reflect**.
3. Add general feedback if it will help students review their response.
4. Set the default grade and configure the response environment in **Advanced settings**.
5. Tag the relevant learning outcome or outcomes and select **Save**.
6. Preview the question in the LAMS activity where students will answer it.

### Write an effective essay prompt

* Ask a focused question rather than naming a broad topic.
* Make the required intellectual task explicit.
* Define the relevant context, audience, role or case.
* State whether students must use sources, examples, data, course concepts or personal experience.
* Identify any required structure, method, referencing style or code constraints.
* State the expected response length and available time where relevant.
* Avoid hidden requirements that appear only in the marking criteria.
* Check that students can answer the prompt within the available time and word range.
* Avoid prompts so broad that marking becomes a judgement of which question the student chose to answer.
* Ensure that the language does not create unnecessary barriers unrelated to the outcome.

An authentic prompt can ask students to produce a response for a meaningful disciplinary or professional context. Authenticity should strengthen alignment, not add decorative role-play or irrelevant complexity.

### Plan marking before deployment

Create a rubric or marking guide that identifies the qualities you will judge and how they contribute to the grade. Criteria may include:

* accuracy and relevance of knowledge;
* quality of analysis or reasoning;
* use and evaluation of evidence;
* coherence of argument or solution;
* application to the specified context;
* originality or insight, where appropriate;
* disciplinary communication; and
* technical correctness, readability or testing for code.

Describe observable qualities rather than vague labels such as **excellent** or **poor**. Decide whether language accuracy is an assessment criterion; do not allow it to influence marks informally when it is not part of the learning outcome.

For consequential assessment:

* provide markers with the same criteria and examples;
* agree how to handle unexpected but valid approaches;
* use calibration or moderation where more than one marker is involved;
* consider anonymous marking where appropriate; and
* record the reasons for grade changes.

The essay editor does not display a rubric field in the screen shown. Provide the criteria through the relevant LAMS activity, accompanying instructions or institutional assessment process so students and markers can access them.

### Add feedback for students

Enter **General feedback** that will be useful after students submit their response. You can include:

* an indicative answer or response outline;
* key principles that a strong response should address;
* an annotated example;
* common misconceptions or weaknesses;
* questions students can use to evaluate their work; or
* resources for the next stage of learning.

General feedback is shared across students and does not replace individual feedback on their response. Avoid presenting one model answer as the only valid approach when the task permits several defensible arguments or solutions.

Check the settings of the activity that uses the banked question, because that activity determines when feedback appears and how individual marking feedback is delivered.

### Configure the advanced settings

#### Default question grade

Enter the maximum mark or weight for the question. The interface accepts an integer, for example `1`, `5` or `10`.

Set a value that reflects the scope, difficulty and time required. Ensure that your marking criteria add up to this value. A heavily weighted essay should provide correspondingly substantial evidence of important learning outcomes.

#### Code style

The default selection is **None**. Keep it when students will write ordinary prose.

Select an available code style when students must submit program code and the response editor should present it appropriately. Choose the style that matches the language or notation required by the task.

Code styling improves presentation; it does not run, compile, test or validate the code. If the outcome requires executable behavior, testing, multiple files or a development workflow, use an assessment environment designed to capture that evidence.

Tell students whether you will assess syntax, correctness, efficiency, documentation, test coverage or explanation. A code-only response may not reveal the student's design reasoning.

#### Allow students to use rich text editor

Turn this on when students need formatting tools such as headings, lists, links, tables, equations or embedded media. The screenshot shows this setting enabled.

Turn it off when plain text is sufficient or when formatting could interfere with code or structured text. Do not require elaborate presentation unless it contributes to the learning outcome.

Preview the editor with keyboard navigation and assistive technology where possible. Make sure students can produce every required element without relying on inaccessible formatting.

#### Template

Enter content that students should receive in their answer field when they begin the question. A template can provide:

* section headings;
* a response framework;
* prompts for required evidence;
* a table to complete;
* starter code or a function signature; or
* a structure for reflection or case analysis.

Use a template to clarify the task and reduce irrelevant organisational load. Do not over-structure the response when planning and organisation are part of what you intend to assess.

Keep instructions in the question description. Use the template for content students should complete, retain or modify. Preview it carefully so placeholder text cannot be mistaken for part of the student's answer.

#### Maximum number of words

Select **Maximum number of words**, then enter the upper limit when you need to constrain response length.

Set a limit that allows students to demonstrate the learning outcome without rewarding unnecessary length. Tell students whether headings, quotations, citations, tables or code count towards the limit, and check how LAMS counts them.

#### Minimum number of words

Select **Minimum number of words**, then enter the lower limit when a minimum is necessary.

Use minimum limits sparingly. Word count is not evidence of quality, and students may add irrelevant text merely to satisfy the setting. A clear account of the required evidence is usually more useful than a high minimum.

If you use both limits, ensure the range is realistic and test the editor's word-count behavior before deployment.

#### Learning outcomes

Search by outcome name or code, then select every learning outcome for which the essay provides meaningful evidence.

Tag the outcome at the level actually assessed. A prompt that asks students to **describe** a concept does not provide evidence for an outcome that requires them to **critically evaluate** or **design**. Ensure the prompt, marking criteria and outcome use compatible expectations.

Avoid tagging every broadly related outcome. An essay may address several topics while providing strong evidence for only a small number of outcomes.

### Design essays in an AI-enabled environment

Assume students have access to generative AI outside controlled settings. Make the rules explicit rather than leaving students to infer them.

Depending on the learning outcome, you may:

* prohibit AI assistance for a clearly explained reason;
* allow limited uses such as brainstorming or language support;
* require students to critique or improve AI-generated material;
* require an acknowledgement of tools, prompts and changes; or
* assess process evidence such as notes, drafts, source decisions, feedback responses or a short oral follow-up.

Do not redesign the task solely as a contest to produce something an AI system cannot generate. Current quality guidance favours authentic, inclusive and AI-aware assessment that makes human learning visible and aligns the use of technology with the intended outcome ([TEQSA, 2026](https://www.teqsa.gov.au/guides-resources/protecting-academic-integrity/academic-integrity-toolkit/risks-academic-integrity-ai/adapting-assessment-age-generative-ai-assessment-adaptation-model); [QAA resources](https://www.qaa.ac.uk/en/sector-resources/generative-artificial-intelligence/qaa-advice-and-resources)).

Apply institutional policy consistently and tell students how permitted assistance affects the marking criteria. Do not infer misconduct from writing style alone.

### Use essays within a programme of assessment

An essay can provide rich evidence, but it is still one sample produced under particular conditions. Combine essay evidence with other assessment methods when outcomes include oral communication, collaboration, practical performance or sustained professional practice.

Use lower-stakes drafts, peer feedback or staged submissions to support learning before a consequential final response. In a programmatic approach, feedback from each assessment event should inform future learning, while higher-stakes decisions draw on multiple data points rather than one essay alone ([van der Vleuten et al., 2012](https://pubmed.ncbi.nlm.nih.gov/22364452/)).

### Review the question after use

* Check whether students interpreted the prompt as intended.
* Identify requirements that markers applied but the prompt did not state.
* Review the distribution of marks by criterion, not only the total grade.
* Compare marker judgements and investigate substantial disagreement.
* Check whether the word limits supported concise evidence or distorted responses.
* Review whether the template scaffolded students appropriately.
* Examine whether AI-use expectations were clear and workable.
* Revise ambiguous criteria, inaccessible formatting or unnecessary workload before reuse.
* Record substantive changes as a new version so results from materially different prompts are not treated as directly comparable.

### Final check

Before you save or deploy the question, confirm that:

* the prompt measures a tagged learning outcome at the intended level;
* students know what they must produce and what evidence to use;
* the marking criteria are available and aligned with the maximum grade;
* the code style and editor match the required response;
* the template supports rather than answers the task;
* word limits are necessary, realistic and clearly explained;
* permitted and prohibited uses of AI and other resources are explicit;
* marking, moderation and individual feedback processes are ready;
* the question and response editor are accessible; and
* you have previewed the complete student experience in the receiving LAMS activity.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.lamsfoundation.org/lams/question-bank/essay.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.