Skip to main content
Evaluator last updated August 27, 2026.

Overview

The Reference Knowledge Demands evaluator assesses how much a text’s meaning depends on relational knowledge – i.e. what a text assumes students have already encountered (from another text, a cultural reference, genre conventions, etc.) and then builds on to create meaning. Reference knowledge demands are one of the least visible qualitative dimensions of text complexity – unlike Vocabulary Complexity or Sentence Structure, there is no word to flag or clause to count.
Reference Knowledge Demands is distinct from Background Knowledge Demands:
  • Background Knowledge Demands – Content that a text assumes (facts, concepts, domain knowledge)
  • Reference Knowledge Demands - Relational knowledge that a text assumes (prior encounters with texts, references and conventions)

At a glance

The evaluator was built and validated using the model and temperature below (other configurations will produce different results and may have lower accuracy):

Getting started

Follow the Quickstart to start using this evaluator:

Inputs

Output

Interpreting results

High or low complexity scores are not inherently good or bad. A higher complexity score simply indicates a text whose meaning depends more substantially on prior cultural or textual encounters.
Example: A “Very complex” or “Exceedingly complex” text may be ideal if the instructional goal is analyzing how authors use allusion and convention to build arguments. However, that same text may not be a good fit if the instructional goal is content comprehension.

Evaluator release history

Literacy evaluators

Explore other evaluators that assess qualitative text complexity.

Literacy dataset

Explore the expert-annotated benchmark behind literacy evaluators.

Quickstart

Run an evaluator in the Evaluators playground, a Python notebook, or with the SDK.