A second grader who can read help has a head start on helper, helpful, helping, helpless, and unhelpful. Each new word adds a familiar part to a word she already knows, and the meaning of the whole follows from its parts. She doesn’t need to learn six separate words. She needs to learn one family and a few prefixes and suffixes that show up again and again.
That’s why the morphological family, rather than the single word, is the right unit for vocabulary instruction once students have foundational skills.. Nagy and Anderson (1984) estimated that printed school English contains about 88,500 word families. That number is too large to guide teaching. The more useful question is how many families do most of the work in the texts students read.
Amanda Goodwin, Gina Cervetti, and I took up that question several years ago (Hiebert et al., 2018). When we sorted the most frequent words in school texts into families, we found 2,451 families, averaging four to five members each. Most of them (85%) appear by the end of the elementary grades. Our analysis had two limits, though. It covered only a slice of English, the roughly 19,500 most frequent words in school texts. And the families we identified were never released as a resource that teachers and researchers could search and use.
Clarence Green of the University of Hong Kong has given us a way past both limits. Clarence developed the British and American Word Families database (BAWF), which codes the base word, roots, and affixes of about 378,000 words from large corpora of British and American English. I’ve had the pleasure of collaborating with Clarence and with Averil Coxhead of Victoria University of Wellington on the first study describing BAWF. Our peer-reviewed, open-access article, led by Clarence, has just been published in Frontiers in Psychology (Green et al., 2026). It describes how the database was built, how its coding compares with established databases created by human analysts, and what BAWF reveals about the words in children’s texts. Anyone who wants the full story of the database’s creation can read the article for free.
Using BAWF, we estimated that about 2,100 to 3,100 familiar word families account for 90% of the running words in school-aged texts. Two very different methods, years apart, point to the same answer: a few thousand families give students access to most of what they read.
BAWF also shows how morphology grows across the grades. In U.S. core reading programs, words with more than one meaningful part make up about 17% of the different words in grade 1 and about 37% by grade 6. Words built on Greek and Latin roots appear even in picturebooks. And the 100 most productive prefixes and the 100 most productive suffixes account for about 94% and 95% of the words that carry them. For teachers, that’s a manageable list.
Not every family is as tidy as help. Nation and national share meaning but not sound. Busy and business share spelling, but the meaning link has mostly faded. Students benefit when teachers point out the connections that hold and show them how to check a guess about word parts against the context. Ratings of how transparent each family is are among the next additions planned for BAWF.
We’re pleased to make BAWF available through TextProject. Teachers, curriculum developers, and researchers can search word families, filter by prefix, suffix, or root, sort by frequency and grade level, and build their own word lists. The database is free and open, and users can report parses they think need a second look, so BAWF will keep improving.
Explore BAWF: https://textproject.org/word-families-project
Read the underlying research: To read the study of how the database was built and validated, see the open-access, refereed article: Green, C., Hiebert, E. H., & Coxhead, A. (2026). Patterns of morphology in children’s print: Estimates of polymorphemic vocabulary using an innovative, AI-informed database of British and American word families. Frontiers in Psychology, 17, Article 1985734. https://doi.org/10.3389/fpsyg.2026.1985734
References
Green, C., Hiebert, E. H., & Coxhead, A. (2026). Patterns of morphology in children’s print: Estimates of polymorphemic vocabulary using an innovative, AI-informed database of British and American word families. Frontiers in Psychology, 17, Article 1985734. https://doi.org/10.3389/fpsyg.2026.1985734
Hiebert, E. H., Goodwin, A. P., & Cervetti, G. N. (2018). Core vocabulary: Its morphological content and presence in exemplar texts. Reading Research Quarterly, 53(1), 29–49. https://doi.org/10.1002/rrq.183
Nagy, W. E., & Anderson, R. C. (1984). How many words are there in printed school English? Reading Research Quarterly, 19(3), 304–330. https://doi.org/10.2307/747823