Early Release

This evaluator reflects early-stage work. We’re continuously improving its accuracy and reliability.
Take a look at the evaluator files.

​
What it is

This evaluator helps assess how challenging students may find the vocabulary of AI-generated texts aimed at Grades 3-4.

​
Why this matters

Vocabulary is the strongest predictor of reading comprehension, but existing readability metrics barely scratch the surface of its complexity. Words vary in familiarity, specificity, and academic utility, Tier 2/3, and these differences can make a text either accessible or impenetrable for students. The Vocabulary Evaluator gives developers the fine-grained insight they need but can’t get from traditional tools. It helps determine whether texts use words that align with grade-level expectations and support growth in academic language. This ensures students are consistently exposed to the kinds of vocabulary that build knowledge and enable them to fully engage with grade-level texts. By understanding what makes a text difficult for a student to read, edtech companies and educators are better equipped to ensure students get the right text for their needs, along with the right instructional supports. You can use this evaluator to help ensure AI-generated texts are sufficiently complex for the grade level and their intended purpose. For example, experts have taught us that:
  • The complexity of the texts students work with should increase across the year.
  • Anchor texts should be rich and complex (and increasingly so across the year).
  • Supplementary texts may be intentionally simpler if they are aimed at scaffolding students’ background knowledge on a topic in the anchor text rather than working with vocabulary.

​
How it works

When educators decide if vocabulary will be challenging for their students, they first think about what background knowledge the students likely have. This evaluator works similarly, breaking the evaluation into two steps:
  1. It estimates a student’s background knowledge given the selected grade level.
  2. It uses the background knowledge estimate as a starting point to evaluate the complexity of a passage’s vocabulary.