The Development of Phonological Awareness in Preschool Children (2003)

J. M. Carroll, M. J. Snowling, C. Hulme & J. Stevenson. (2003). Developmental Psychology 39(5), 913-923.  DOI: 10.1037/0012-1649.39.5.913

Longitudinal study of preschool children's development of phonological awareness.

A total of 67 3- and 4-year-old children were tested three times within a year on a range of phonological
awareness tasks. Forced-choice tasks measuring syllable, rime, and initial phoneme awareness were used. Goswami and Bryant’s theory (1990) predicts that the syllable matching task will be easiest for children while rime and initial phoneme tasks will be equally difficult because they both require sensitivity to the onset–rime, a more complex level of phonological awareness.

Lexical restructuring theory proposes that children begin to develop phonological awareness incrementally, and that children’s phonological awareness is tied to the development of "increasingly
segmentalized lexical phonological representations." First they become aware of entire words, then they develop the representation of the sounds within words. Since PA tasks measure a child’s knowledge of the individual sound units within words, it follows that PA is highly dependent on the development of a child’s lexical representations.

This study examined three research questions:
1. Does the development of phonological awareness follow the progression of syllable, then onset and rime, then phoneme?
2. Are language skills such as vocabulary, detection of mispronunciations, and articulation related to the development of phoneme awareness?
3. What role does letter knowledge play in the development of phonological awareness? (p. 915)

Procedures

Time 1, (age avg = 3 yr 10 mo) the letter knowledge, vocabulary, and phonological matching tasks given.
Time 2, (4 months later) same tasks and two additional tasks: mispronunciation detection and articulation. 
Time 3, (8 months after T2 age 4 yrs 9 mo) All prior tasks were given again (except for syllable matching)  and explicit phoneme awareness tasks added: syllable matching first, rime matching, then initial-phoneme matching tasks. All PA tasks presented in separate sessions.
Language tasks were interspersed with PA tasks, consisting of 1) a mispronunciation task (used a puppet to mispronounce), 2) receptive language (British picture vocabulary knowledge), and 3) an articulation task scoring the percentage of correct phonemes pronounced by children identifying pictures by naming them (some large polysyllabic words like caterpillar).

Results indicate that children detected syllables at higher levels than phonemes and children had learned many of the alphabet letter names by the end of the testing period.

Conclusions: "It was clear in the present study that children tended to develop syllable and rime awareness before phoneme awareness. However, there was little sign of any difference in levels of performance between syllable and rime awareness tasks. It is also notable that in our confirmatory factor analysis of data from
Time 2, both syllable and rime awareness loaded highly on a single underlying latent variable (large-segment awareness). It seems, therefore, that a better way of characterizing development might simply be as a progression from awareness of large units (syllables and rimes) to awareness of small units (phonemes). This characterization is broadly in line with Gombert’s (1992) separation between epilinguistic and metalinguistic phonological awareness (p. 920)."

Further comments worth noting: "It should be noted that in line with a number of previous studies, the present one provides evidence that rime awareness and phoneme awareness are separable skills (920)."


"Implicit, large-segment sensitivity is a skill that grows out of normal language development. In fact, it seems to
interact closely with receptive lexical knowledge and might therefore be better considered a part of normal linguistic, rather than of metalinguistic, development. The later development of the explicit awareness of phonemes appears to build on the foundation of earlier large-segment awareness and to depend, in addition, on the accuracy of a child’s articulation skills. We propose that the development of these two types of phonological awareness reflects the development from global to segmental phonological representations." (922)


Thoughts on this study: No type of instruction was provided within the study to induce learning but children were tested over a year's time to see if they "naturally" developed PA in the course of a typical preschool curriculum. Since we know absolutely nothing about the learning environments of the children, we really can't comment with any degree of certainty on the progression of levels of "phonological sensitivity."

My questions remains unanswered after looking at the evidence in this study: If preschool children are given explicit directions to help them successfully identify phonemes, is instruction in syllable detection necessary? Can children infer other concepts of print through phoneme identity instruction? Will phoneme identity instruction initiate an understanding of the alphabetic principle in preliterate children? Is it necessary for them to begin to apply their knowledge in literacy tasks and what might these be without formal reading instruction?