Sentence Structure

Evaluator last updated February 18, 2025.

Overview

The Sentence Structure evaluator assesses the complexity of sentence structure in informational texts relative to a specified grade level:

Identifies sentence features in the text, including sentence type composition, average words per sentence, subordinate clause ratios, and concepts per sentence.
Assigns an overall complexity rating using an LLM, combined with statistical thresholds for sentence features.

At a glance


Input type	Informational text
Supported grades	3–12
Rubric	SAP ↗‘s Qualitative Text Complexity Rubric for Informational Text ↗

The evaluator was built and validated using the model and temperature below (other configurations will produce different results and may have lower accuracy). The evaluator must be run in two stages — combining them into a single step reduces accuracy.


Model used	GPT-4o
Temperature	0

Getting started

Follow the Quickstart to start using this evaluator:

Access method
Evaluators Playground	View in the Learning Commons Platform ↗
SDK	TypeScript ↗
Python notebook	View in GitHub ↗
Prompts	View in GitHub ↗

Inputs

Input	Description	Required
Target grade level	Enables grade context evaluation	Yes
Text type	Informational text optimal length: 100–200 words (max ~1,200 characters)	Yes

Output

Field	Description
Answer	Complexity score (“Slightly complex”, “Moderately complex”, “Very complex”, or “Exceedingly complex”) See Interpreting results for grade-specific criteria
Reasoning	Explanation of the rating based on identified sentence features and grade context

Interpreting results

Grade 3

Complexity score	How to use
Slightly complex	Simple, straightforward language and sentence structures Meets 2+ of the following criteria: Sentence type: Typically > 60% simple sentences Sentence length: Typically < 12 average words per sentence Subordination: Typically < 25% of sentences have subordinate clauses
Moderately complex	Mix of simple and more complex sentences: Sentence type: Typically 40 - 60% simple sentences Sentence length: Typically 12 - 16 average words per sentence Subordination: Typically 25 - 45% of sentences have subordinate clauses
Very complex	More elaborate sentences with multiple clauses and ideas: Sentence type: Typically < 40% simple sentences Sentence length: Typically 16 - 19 average words per sentence Subordination: Typically > 45% of sentences have subordinate clauses
Exceedingly complex	Dense with very long, intricate sentences and high subordination Meets 2+ of the following criteria, including 1+ from “Structural density”: Structural density Subordination: Typically > 50% of sentences have subordinate clauses Multiple subordination: Typically > 12% of sentences have 1+ subordinate clause Syntactic complexity: Typically > 15% compound-complex sentences Length Sentence length: Typically > 19 average words per sentence High concentration of very long sentences: Typically > 15% of sentences have 30+ words

Grade 4

Complexity score	How to use
Slightly complex	Simple, straightforward language and sentence structures Meets 2+ of the following criteria: Sentence type: Typically > 55% simple sentences Sentence length: Typically < 13 average words per sentence Subordination: Typically < 30% of sentences have subordinate clauses
Moderately complex	Mix of simple and more complex sentences: Sentence type: Typically 40 - 55% simple sentences Sentence length: Typically 13 - 17 average words per sentence Subordination: Typically 30 - 50% of sentences have subordinate clauses
Very complex	More elaborate sentences with multiple clauses and ideas: Sentence type: Typically < 40% simple sentences Sentence length: Typically 17 - 22 average words per sentence Subordination: Typically > 50% of sentences have subordinate clauses Multiple subordination: Typically > 8% of sentences have 1+ subordinate clause
Exceedingly complex	Dense with very long, intricate sentences and high subordination Meets 2+ of the following criteria, including 1+ from “Structural density”: Structural density Subordination: Typically > 60% of sentences have subordinate clauses Multiple subordination: Typically > 15% of sentences have 1+ subordinate clause Syntactic complexity: Typically > 20% compound-complex sentences Length Sentence length: Typically > 22 average words per sentence High concentration of very long sentences: Typically > 15% of sentences have 30+ words

Grades 5-12

Complexity score	How to use
Slightly complex	>= 50% simple sentences Exception: Moderately Complex if >= 50% simple sentences AND >= 20% compound sentences Important: NEVER includes advanced complex sentences
Moderately complex	Primarily simple and compound sentences, with some complex constructions; can take on any distribution of sentence types as long as there aren’t > 2 advanced complex sentences and as long as there aren’t so many simple sentences that the text becomes Slightly Complex. May contain many simple sentences, compound sentences, and/or basic complex sentences, as well as 1 - 2 advanced complex sentences.
Very complex	3+ advanced complex sentences Exception: Exceedingly Complex if >= 65% advanced complex sentences
Exceedingly complex	>= 65% advanced complex sentences

Accuracy and validation

This evaluator is provided as Early access. Comprehensive accuracy measures are not yet available. Validation testing is ongoing.

We assessed performance against an expert-annotated dataset of CLEAR Corpus ↗ texts — ~480 for Grade 3 and ~480 for Grade 4. Accuracy has been most extensively validated on Grades 3–4. We are still evaluating the performance for grades 5-12.

Metric	Description	Result
Expert agreement	The percentage of evaluated examples where at least one expert agreed with the evaluator’s rating during review testing.	53% agreement for Grade 3 54% agreement for Grade 4
Accuracy within one-level	The percentage of evaluator ratings that fall within the same complexity level or one level away from the of expert annotators’.	94%
Baseline comparison	How the evaluator’s accuracy compares to a simple, unrefined prompt.	26% more accurate

For more information on our validation process, see Accuracy.

Evaluator release history

Date	Changed
February 18, 2026	Added grades 5-12.
September 23, 2025	First release

Understanding evaluators

Getting started

SDK API Reference

Literacy evaluators

Feedback evaluators

Standards evaluators

Datasets

Resources

Overview

At a glance

Getting started

Inputs

Output

Interpreting results

Grade 3

Grade 4

Grades 5-12

Accuracy and validation

Evaluator release history

​Overview

​At a glance

​Getting started

​Inputs

​Output

​Interpreting results

​Grade 3

​Grade 4

​Grades 5-12

​Accuracy and validation

​Evaluator release history

Overview

At a glance

Getting started

Inputs

Output

Interpreting results

Grade 3

Grade 4

Grades 5-12

Accuracy and validation

Evaluator release history