Having reviewed freely accessible research I found references to The Grapheme-Phoneme Problem in Reading and other spelling studies and have sought other frequency tables that describe the relative frequency of the letters (graphemes) used to represent the sounds (phonemes). There are available online resources for frequency tables of words in spoken language and printed texts, but not as readily available are comparable tables for the graphemes and/or phonemes used in UK English.

As a primary school teacher, I am curious as to whether it would be worthwhile focussing on teaching children those graphemes that occur more frequently, as is done with high frequency sight words. This question may be of use to those professionals that support children with decoding when reading and using phonics knowledge when spelling.

I’ve done my own diagnostics for spelling errors but I’m no statistician and so am seeking help from those that may have access to relevant publications and studies.

  • 2
    Printed texts don’t have phonemes, so that bit is not answerable. Jan 9, 2019 at 23:03
  • Connie, thank you for the additional detail. You may want to edit the question title as well, though.
    – Andrew Leach
    Jan 10, 2019 at 23:03

2 Answers 2


I found two resources focused on British spelling that deal with frequency.

The Dictionary of the British English Spelling System (2015) is available through the publisher in PDF for free. It directly tackles the grapheme/phoneme problem and diagrams the possible grapheme combinations for each phoneme. While there isn't a simple list of graphemes in order of frequency, chapter 3 gives frequencies of graphemes for each phoneme in consonants, and chapter 5 does the same for vowels. Meanwhile, chapter 8 gives comprehensive lists of graphemes/phonemes, and Appendix B ("Pedagogically selected lists of phoneme-grapheme and grapheme-phoneme correspondences") is both more accessible and groups phonemes and graphemes according to its "basic" (read: most common) form and its "other" variants.

Then there's this article: Gontijo, P.F.D. et. al., "Grapheme-phoneme probabilities in British English." Behavior Research Methods, Instruments, and Computers 35.1 (2003): 136-157. The article is pretty technical, but its appendix is worth a sit-down over tea: for each grapheme they generated the probability that it would represent a given phoneme. If you just look at the word that it uses as an example, it'll give you a good idea of how each grapheme is used. For instance, "f" is used to represent /f/ about 2/3rds of the time. (The other third of the time it represents /v/, as in "of.")

I hope those help.

  • This is a resource I didn’t know about. I’ll review the content throughout the next week. Jan 13, 2019 at 21:57

From various sources I could find and doing my own calculations, 41% of the letters in British English texts are vowel letters (counting Y in words like "bay" but not in "yard"). The large number of silent vowel letters means that counting phonemes reveal that only 39% of them are vowels. That number appears to be closer to 37% in American English, due to the abundance or rhotics.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.