Best predictors of early reading achievement

Researchers studied six pre-reading skills to determine which predicted reading achievement at the end of first and second grades. Darrell Morris and Janet Bloodgood, Appalachian State University, and Jan Perney, National-Louis University/Chicago, assessed the pre reading skills of 102 Children during kindergarten and their reading skills at the end of first and second grade. They found that four kindergarten skills–alphabet recognition, word recognition, concept of word in text (ability to point to a word as it is read), and spelling of beginning and ending consonants–effectively predicted reading ability in the primary years. Accurate predictions could be made by midyear. Surprisingly, the phonemic spelling task was superior to the more commonly used oral phoneme segmentation task in predicting beginning reading .

These researchers tested children’s pre reading skills at the beginning, middle and end of the kindergarten year. At the three points during kindergarten they tested alphabet recognition, word recognition, oral awareness of beginning consonant sounds, concept of word in text, spelling with beginning and ending consonants, and oral phoneme segmentation.

Different skills were predictive at different times during the kindergarten year, the research found. However, two skills, alphabet recognition and concept of word in text, predicted first-grade reading achievement throughout the kindergarten year. Oral phoneme segmentation was not a significant predictor of later reading achievement at any point during the kindergarten year. Two seldom-used skills–concept of word in text and phonemic spelling–were better predictors than two widely used variables–beginning consonant awareness and phoneme segmentation. Morris et al. suggest that concept of word in text captures a child’s emerging phonological skill and letter/sound knowledge. Until a child can accurately point to words as they are read (match spoken to printed words), he or she will have difficulty developing a sight vocabulary or using letter-sound cues to decode words. Spelling with first and last consonants was a significant predictor, in these researchers’ opinion, because it involves a type of phoneme segmentation but is easier than oral segmentation and more sensitive to early literacy development.

Limitations and Implications of Study

These researchers point out two limitations to the generalizability of these findings. First, all students in this study received a combination of letter/sound work, supported contextual reading, and writing with invented spelling in their kindergarten curriculum. A different instructional emphasis may produce a different set of predictors. Second, the sample of children included in this study came from rural, lower-middle class families and most did not attend preschool. On entering kindergarten, children in this study, on average, recognized less than half of the alphabet letters. In other populations predictors of reading might vary.

Morris et al. report that by the middle of kindergarten a small number of pre reading skills accurately predict early reading. These include alphabet recognition, word recognition, concept of word in text, and spelling with beginning and ending consonants. Developmental spelling ability in kindergarten is a surprisingly effective predictor of first-grade reading achievement and clearly superior to oral phoneme segmentation in predicting achievement. Spelling appears to be an easier way for pre readers to demonstrate their phoneme awareness. These researchers recommend substituting it for the common oral phoneme task. Moreover, performance on just two middle-of-kindergarten tasks (alphabet knowledge and spelling) correctly predicted early reading skills 85 percent of the time. Screening these pre reading skills midyear can identify children at risk of failure so they can be given extra help to improve their literacy skills.

“Kindergarten Predictors of First- and Second-Grade Reading Achievement”, The Elementary School Journal, Volume 104, Number 2, November 2003, pp. 93-110.

Published in ERN December/January 2004 Volume 17 Number 1

Leave a Reply

  • (will not be published)