Embedded Phonics

In a two-year study of Embedded Phonics (EP), Torgesen and his colleagues (2001) demonstrated large and lasting effects for students reading well below grade level. The content of instruction in EP is individualized for each student, but the pattern of instructional activities is consistent. One-to-one instruction is divided into two 50-minute sessions each day for eight weeks. The first session each day has five activities for sight words, phonics, and guided reading in context:

  1. Practice in Reading Sight Words (Ten Minutes). The student decodes words from a word list and identifies those parts that “play fair”--follow familiar phonics rules—and those that do not. Each word comes with a phonemic guide for scaffolding. The guide is removed as the student more readily identifies the word. Words are practiced over multiple trials until student decodes them within one second over three successive days.
  2. Spelling Newly Introduced Sight Words (Five Minutes). The student “stretches” words and listens for all their sounds while spelling to learn spelling patterns and phonemic awareness. The student learns to memorize the irregular parts of words that do not play fair.
  3. Word Games for Fluency with Sight Words (Ten Minutes). The student identifies sight words in the course of playing games.
  4. Phonics Mini lesson (Ten Minutes). The student learns the rules and skills of phonemic decoding and spelling. Rules include common spelling variations of the consonant and vowel phonemes, signal (“silent” or “final”) e, and r-controlled vowels. Skills include blending sounds together to form words, adding inflected endings, and analyzing syllable patterns.
  5. Oral Reading in Trade Book or Basal Series (Fifteen Minutes). The student practices reading in either a graded series of trade books or a basal series with controlled vocabulary. The student reads words accurately but focuses on meaning. Word-level errors are corrected in two ways: If the error involves a phonemic decoding principle that has been taught, the student compares the phonemes he used to the word as written; if the error does not make linguistic sense in the context of the passage, the student considers whether the word he said made sense. The student then sounds out the word to produce a word that “sounds like that and makes sense in the sentence.” As the student reads, the teacher asks specific questions, has him summarize the passage to that point, and has him predict the next part of the passage.

The second 50-minute session had four activities for sight words, spelling, reading in context, and writing in context:

Intensive instruction with this program was followed by large improvements the reading scores of many students who had struggled to read prior to treatment. Particularly noteworthy is the change in mean standard score on a passage comprehension test. Before treatment the mean standard score was 82.2 (S.D.=11.0), after treatment it was 92.0 (S.D.=19.8), and two years later it was 96.9 (S.D.=11.5). This is a gain of nearly one standard deviation with an upward trend two years after the end of treatment. The authors’ methodological choices limit the conclusions possible from this study, but within those limits, the results are impressive.

home diagram

Torgesen et al, 2001