Friday, May 27, 2011

Gaining Alphabetic Insight: Is Phoneme Manipulation Skill or Identity Knowledge Causal? (1998)

Murray, B. A. (1998) Journal of Educational Psychology, 90(3), 461-475.

Purpose of study: Phoneme Awareness often refers to an ability to manipulate and blend phonemes. This study looks at it from another angle and seeks to confirm Byrne and Fielding-Barnsley's findings that phoneme identity is the crucial task beginners need to develop the alphabetic principle and apply it in word identification.

Regarding the findings of Byrne & Fielding-Barnsley (1990) ". . . phoneme segmentation showed a weaker relationship to phonetic cue reading than did identity. Identity scores (r = .49) were better predictors of performance in phonetic cue reading than were segmentation scores (r = .20). Byrne and Fielding-Barnsley (1990) concluded that phoneme identity training is more successful than segmentation training in leading children toward the alphabetic insight because identity was easier to teach and led to a more stable alphabetic insight" (p. 463).

The Problem under investigation: "Do children better learn about the phonemic structure of words
through instruction in generalized manipulation skill, through instruction in particular phoneme identities, or through indirect language experiences? Do children more readily begin using the alphabet to decode when they learn to manipulate phonemes or when they learn the identities of particular phonemes?
Participants: 48 children in kindergarten (5 classrooms 2 parochial, 3 public); 15 day treatments
  • Word identification. Preprimer word list and passage from Basic Reading Inventory (BRI), Fifth Edition, Form A (Johns, 1991) for screening out readers with more than 3 preprimer words correct. 
  • The test of phonetic cue reading (TPCR), (Experimenter design based on Byrne and Fielding-Barnsley (1990)
  • Oral vocabulary. Peabody Picture Vocabulary Test, Form L (PPVT; Dunn & Dunn, 1981).
  • Alphabet knowledge 
  • Phoneme awareness. phoneme manipulations (TPM), an experimenter-constructed adapted from measures used in earlier studies (Stahl & Murray, 1994)
  • The Test of Phonological Awareness (TOPA; Torgesen & Bryant, 1994), modified for the study. 
  • Phoneme awareness.
    • an alternate form of the TPM 
    • the Word-to-Word Matching Test (WWMT) from Gunning (1992), modified 
  • Reading. 
    • Count of trials to mastery in learning eight letter-phoneme correspondences, (maximum of 20 trials).
    • An alternate form of the TPCR.
    • Experimenter-constructed test of decoding, featuring two- and three-phoneme words using the eight letters for taught correspondences with simple one-to-one relationships between phonemes and letters.
    • Experimenter-constructed attitudinal measure, using a Likert scale of images of the cartoon character Garfield (McKenna & Kear, 1990).
Conditions: Three instructional treatments involved intervention programs for phoneme identities (PI), phonological manipulation (PM), and language experience (LE).
  • PI learned a limited number of phoneme-grapheme correspondences. Practiced them in tongue twisters, used memorable activities, and practiced finding them in words in initial and end positions. Also blended and segmented the target phonemes. Phoneme-grapheme correspondences were not taught in this condition. Oral only.
  • PM manipulated phonemes through blending and segmentation, first as onset and rime activities then using the complete phoneme sequence. 
  • LE engaged in developmentally appropriate early literacy activities: looked at storybook  illustrations, listened to stories read aloud, talked about stories, and created a group story. Stories were typed and used as a text the next day.
  • All participants were individually taught the phoneme-grapheme correspondences for eight letters  (F, L, M, N, S, T, E, A) in their final session (Day 15) using a paired-association method.
"[W]hen instruction emphasizes phoneme manipulations, children learn what they were taught. By
contrast, teaching beginners about phoneme identities does not seem to enhance phoneme manipulation skill. This suggests that phoneme manipulation skill may be relatively independent of the knowledge of phoneme identities. Consistent with this explanation, the gains made by the manipulations group did not translate into alphabetic insight on the TPCR" (p. 470).

The PI group failed to show evidence of the superiority of the treatment over the manipulation and segmenting group, however, the differences between groups "emerged at posttesting on the TPCR, a
measure presumed to require phoneme identity knowledge" (p. 470).


The identity group failed to acquire useful knowledge of phoneme identities. The author attributes this to the PI measuring instrument at posttest and limited progress in the field in learning to measure phoneme awareness apart from reading ability. The identity treatment group indirectly demonstrated knowledge of alphabetic insight, but further evidence is needed to conclude that identity knowledge is causal in gaining alphabetic insight. (p. 470)

No comments:

Post a Comment