Phonics and the reading recovery program

Reading Recovery, a remedial program developed in New Zealand by Marie Clay, has been effective in bringing at-risk first-grade readers (the lowest 10 to 20 percent) up to grade level before the end of the year. A child’s reading ability is assessed with a battery of tests Clay developed called the Diagnostic Survey. Students identified as at-risk for poor reading development receive 30 to 40 minutes of individual tutoring by a trained teacher each day as a supplement to reading instruction in the regular class. A child must read at an average or above-average level before tutoring is discontinued.

Each lesson consists of seven activities:  rereading two or more familiar short books; independent reading of the preceding lesson’s new book while the teacher makes a record of errors; letter identification using magnetic letters; writing a story that includes hearing sounds in unfamiliar printed words; cutting up and reassembling the story; introduction of a new book; and reading the new book.

During lessons, phonics are taught as they occur in the context of reading and writing. As a child attempt to spell with the aid of the teacher, he is taught the letter-to-phenome correspondences as needed. The whole process usually takes between 12 and 20 weeks.

Sandra Iversen and William E. Tunmer, Massey University, New Zealand, sought to determine whether the Reading Recovery program would be more effective if systematic instruction in phonics, specifically phonological recoding skills, were incorporated into the program. They compared groups of children in New Jersey who were taught by the standard Reading Recovery program with those who received systematic phonics instruction as part of a modified Reading Recovery program.

Research with poor readers had led these researchers to speculate that such children might be helped more effectively by systematic phonics instruction. Their modified Reading Recovery program did not attempt to teach phonics in an isolated fashion. Phonics instruction replaced the letter-identification segment of the lesson as soon as the child demonstrated he could identify at least 35 of the 54 alphabetic characters. This usually occurred by the fourth week or the sixteenth lesson.

This instruction used common phonograms or word families (“ight” in light, fight and might) because these help to reduce the complexity of vowel-sound generalizations. The aim of the instruction was to make the child more aware that words with common sounds often share spelling patterns.

Choosing a word from one of the books, the teacher first modeled the task and then the child repeated it using magnetic letters to make, break and build new words having similar visual or phonological elements. This work began with manipulating initial sounds/letters/clusters. When the child demonstrated proficiency, the teacher moved on to final and then medial sounds/letters/clusters.


First-grade, at-risk readers were divided into three matched groups of 32 children each: a standard Reading Recovery program, a modified Reading Recovery program, and a standard intervention program in which students received extra help in small groups in a Chapter 1 program. All children were tested with the Diagnostic Survey. Initially, none of the children was able to respond correctly to any of the phonological processing items on the survey tests, demonstrating that these children were particularly deficient in these skills.


Post-testing revealed that compared to the standard intervention group, both Reading Recovery groups scored significantly better on all measures. In general, the results show that the two Reading Recovery groups performed at very similar levels when tutoring was discontinued. The children in the Reading Recovery groups performed as well as and often better than the classroom controls (average and above-average readers in the regular class).

The most significant finding was the difference in the average number of lessons needed before tutoring was discontinued in the two Reading Recovery groups. The average for the modified program was 42 lessons as compared to 57 lessons for the standard program. Therefore, although the Reading Recovery groups performed at very similar levels at discontinuation, the standard group took much longer to reach the same level of proficiency.

This time difference was highly significant and indicates that the standard Reading Recovery program was 37 percent less efficient than the modified program. These results mean that a modified program in which students receive instruction that emphasizes the interrelatedness of sounds and visual patterns shared by different words is more effective. Such a modified Reading Recovery program enables teachers to help more children because it enables children to move back into the regular classroom program much more quickly.

It is important to note that the teachers for the two Reading Recovery programs were trained at different locations and were not aware that some were teaching a modified version of the standard program. Because they were also unaware of the identity of the children targeted for the study, it is unlikely that teacher bias contributed to the large differences between the two groups.

Retesting at the end of the year showed that both Reading Recovery groups continued to perform at very similar levels despite the much briefer intervention period of the modified group.


Iversen and Tunmer conclude that Reading Recovery is a highly effective intervention program, and that one-to-one tutoring is much more effective than typical remedial instruction in small groups. Importantly, however, the teaching of phonic skills only as they occur in the context of reading appears to be less efficient than explicit instruction in the use of letter-to-phoneme correspondences.

The children at risk for poor reading learned to reach much more quickly when given systematic instruction in phonological recoding skills in the modified Reading Recovery program. In particular, a metacognitive approach that exploits the use of phonograms appears to be an effective intervention strategy in the context of the Reading Recovery program.

“Phonological Processing Skills and the Reading Recovery Program”, Journal of Educational Psychology, Volume 85, Number 1, pp. 112-126.

Published in ERN May/June 1993, Volume 6, Number 3

Leave a Reply

  • (will not be published)