It is the most comprehensive resource to date showing how English-speaking children in first through fifth grade read individual words, and what those patterns reveal about early literacy development. The d-ELP is funded by the National Institutes of Health through a Learning Disabilities Hub grant involving UConn, Florida State University, and Hebrew University of Jerusalem. Steacy has been part of the research team since the project began.<\/p>\n \u201cIn the first cycle, we looked at how children navigate English writing from a different angle through the use of small scale experiments,\u201d Steacy says. \u201cThat work showed we needed a large resource that placed words on a common scale, leading directly to this project.”<\/p>\n The team collected isolated word-reading data from about 2,000 children in grades one through five at multiple U.S. schools, intentionally including a broad range of reading abilities and oversampling students at risk for reading difficulties, including dyslexia.<\/p>\n \u201cWe wanted to capture the full range of reading ability,\u201d Steacy says. \u201cIncluding both typical and struggling readers is essential for developing effective tools. Working with students in different contexts allowed us to include a diversity of reading profiles.\u201d<\/p>\n A defining feature of the d-ELP is its use of item response theory (IRT) to estimate difficulty for nearly 10,000 words. Because no child could feasibly read every word, researchers designed assessments that included items below, at, and above each student\u2019s reading level, along with a set of linking items read by all children. This design enabled the team to place every word on one difficulty scale, even though students encountered different subsets of the list.<\/p>\n The resulting dataset provides a scientific infrastructure that did not previously exist, allowing researchers and educators to better match reading materials to student needs. For the first time, they can examine how word difficulty varies across thousands of items and how characteristics such as spelling patterns, the meaningful parts of words (such as roots, prefixes, and suffixes), and the complexity of a word\u2019s sound structure predict children\u2019s reading behavior \u2014 ultimately guiding more effective literacy instruction.<\/p>\n \u201cEnglish has a lot of inconsistency between print and speech,\u201d Steacy says. \u201cThe d-ELP helps us see which words are particularly challenging and why.\u201d<\/p>\n Steacy emphasizes that the project was designed to benefit a wide range of users, including teachers, school leaders, educational publishers, and policymakers. The database is publicly available for noncommercial use, providing all these stakeholders with practical, actionable information to support literacy development.<\/p>\n English has a lot of inconsistency between print and speech. The d-ELP helps us see which words are particularly challenging and why. — Laura Steacy<\/cite><\/p>\n<\/blockquote>\n \u201cWe aimed for a user-friendly resource,\u201d she says. \u201cTeachers can select words by difficulty, spelling, or morphology to plan targeted instruction or assessments. It can support students with specific decoding struggles and help teachers design appropriate materials.\u201d<\/p>\n![\"Laura](\"https:\/\/today.uconn.edu\/wp-content\/uploads\/2025\/12\/Laura-Steacy-300x300.jpg\")
\n