> ## Documentation Index
> Fetch the complete documentation index at: https://docs.learningcommons.org/llms.txt
> Use this file to discover all available pages before exploring further.

# Purpose

> Reference documentation for the Purpose Evaluator.

export const EarlyAccessCallout = ({children}) => <div className="eyebrow-callout not-prose rounded-xl border border-gray-200/80 p-5 dark:border-white/10" style={{
  marginBottom: "1rem",
  borderRadius: "4px"
}}>
    <div className="mb-3">
      <Badge color="green" size="md" icon="flask">
        Early access
      </Badge>
    </div>
    <div className="callout-body text-[15px] leading-relaxed text-gray-700 dark:text-gray-300">{children}</div>
    <style>{`.callout-body a { text-decoration: underline; text-decoration-color: #178251; }`}</style>
  </div>;

[Evaluator last updated May 7, 2026.](#evaluator-release-history)

<EarlyAccessCallout>
  This functionality is actively evolving. Changes may occur as we expand
  capabilities and improve accuracy and reliability. Email
  [support@learningcommons.org](mailto:support@learningcommons.org) ↗ with your
  feedback or issues.
</EarlyAccessCallout>

## At a glance

|                      |                    |
| -------------------- | ------------------ |
| **Input type**       | Informational text |
| **Passage length**   | 200 words or more  |
| **Supported grades** | 3–12               |

A text's purpose is *what it's trying to do*: inform, persuade, explain, describe, or entertain. The Purpose Evaluator assesses how clearly a text communicates its central purpose. It analyzes whether a text's intent is explicitly stated, indirectly hinted at, or masked (e.g., A passage presents itself as neutral information but is really building a persuasive argument).

## Model and prompt

For instructions on running the evaluator, see [Quickstart](/evaluators/getting-started/quickstart).

|                     |                                                                                                               |
| ------------------- | ------------------------------------------------------------------------------------------------------------- |
| **Model used**      | gemini-3-flash-preview                                                                                        |
| **Temperature**     | 0                                                                                                             |
| **Prompts**         | [View prompts](https://github.com/learning-commons-org/evaluators/tree/main/evals/prompts/purpose/) ↗         |
| **Python notebook** | [View notebook](https://github.com/learning-commons-org/evaluators/blob/main/evals/purpose_evaluator.ipynb) ↗ |

The prompt is optimized for the model mentioned above. If you use other models and parameters, the output accuracy and result will vary.

## Inputs

| Requirement            | Supported                                 | Required |
| ---------------------- | ----------------------------------------- | -------- |
| **Target grade level** | Allows grade-specific complexity guidance | Yes      |
| **Text type**          | Informational text                        | Yes      |

## Output

| Field                        | Description                                                                                                                          |
| ---------------------------- | ------------------------------------------------------------------------------------------------------------------------------------ |
| `complexity_score`           | Purpose complexity level: `slightly_complex`, `moderately_complex`, `very_complex`, `exceedingly_complex`, or `more_context_needed.` |
| `reasoning`                  | Overall and detailed explanation of the rating, citing specific text features and their impact on student comprehension.             |
| `adjustment_and_scaffolding` | Suggestions for adjusting the text or scaffolding students to make it appropriate for the target grade.                              |
| `recommended_use_cases`      | Instructional opportunity recommendations for the passage's purpose.                                                                 |

## Interpreting results

The evaluator returns one of the following ratings, along with reasoning, to help you determine your best course of action.

| Rating              | Meaning                                                                                                   |
| ------------------- | --------------------------------------------------------------------------------------------------------- |
| Slightly complex    | The purpose is explicitly stated and concrete. Readers can identify intent without inference.             |
| Moderately complex  | The purpose is mostly clear but may require some inference from context or supporting details.            |
| Very complex        | The purpose is hinted at or subtle. Readers must infer intent from how details are selected or organized. |
| Exceedingly complex | The purpose is masked or layered. The text appears to do one thing while actually doing another.          |
| More context needed | The passage is too short to determine its purpose. Try a longer portion of the text, or a different text. |

More complex ratings indicate a text requires more interpretation by readers. Purpose complexity isn't inherently good or bad; it depends on the instructional goals. A text rated Very complex or Exceedingly complex may be ideal for teaching students to identify implicit or persuasive intent, while the same rating would signal a mismatch if the goal is content comprehension.

<Tip>
  For an overall grade-level judgment that synthesizes all text complexity
  dimensions, see the [Grade-Level Appropriateness
  Evaluator](/evaluators/literacy-evaluators/grade-level-appropriateness).
</Tip>

## Rubric

| Slightly complex                                      | Moderately complex                                         | Very complex                                                                             | Exceedingly complex                                                                           |
| :---------------------------------------------------- | :--------------------------------------------------------- | :--------------------------------------------------------------------------------------- | :-------------------------------------------------------------------------------------------- |
| Explicitly stated, clear, concrete, narrowly focused. | Implied but easy to identify based upon context or source. | Implicit or subtle but fairly easy to infer; more theoretical or abstract than concrete. | Subtle and intricate, difficult to determine; includes many theoretical or abstract elements. |

Based on [SAP](https://learnwithsap.org/)'s ↗ [Qualitative Text Complexity Rubric for Informational Text](https://achievethecore.org/file/823/file-823.pdf) ↗.

## Accuracy and validation

<Note>
  This evaluator is provided as **Early access**. Comprehensive accuracy
  measures are still evolving, and validation testing is ongoing.
</Note>

The evaluator was optimized using 35 annotated passages from the[ CLEAR Corpus](https://www.commonlit.org/blog/introducing-the-clear-corpus-an-open-dataset-to-advance-research-28ff8cfea84a/) ↗ and validated through expert review of additional samples.

| Metric                    | Result                                                                                                     |
| ------------------------- | ---------------------------------------------------------------------------------------------------------- |
| Complexity score accuracy | 84% agreement with expert annotations                                                                      |
| Expert agreement          | 70%                                                                                                        |
| Reasoning soundness       | Average 3.7 / 5                                                                                            |
| Dataset source            | [CLEAR Corpus](https://docs.google.com/spreadsheets/d/1sfsZhhP2umXXtmEP_NRErxLuwgN98TyH7LWOq3j07O0/edit) ↗ |

<Note>
  Exceedingly complex texts aren't common in lower grades and the benchmark
  dataset doesn't include many examples of them. Use caution when applying this
  evaluator to higher grade levels where such texts are more frequent.
</Note>

## Evaluator release history

| Date        | Changes        |
| ----------- | -------------- |
| May 7, 2026 | First release. |