A Stage-Sequential Model of Reading Transitions: Evidence From the Early Childhood Longitudinal Study

Kaplan, D. and Walpole, S. (2005) A stage-sequential model of reading transitions: Evidence from the early childhood longitudinal study. Journal of Educational Psychology 97(4), 551–563.    

Researchers used latent transition analysis to analyze data from the U.S. Gov. sponsored Early Childhood Longitudinal Study (NCES, 2001) to examine reading development during kindergarten and 1st-grade. Data from more than 22,000 children attending 1000 public schools includes poverty status and dichotomous measures of reading at 4 different time periods. In each time period, 5 latent classes were represented: low alphabet knowledge, early phonological processing, advanced phonological processing, early word reading, and early reading comprehension.

Findings indicated in the abstract:

". . . children living below poverty are less likely to experience successful reading transitions than their above-poverty peers. However, children in the below-poverty group who began kindergarten with at least early phonological processing experienced transition probabilities similar to their above-poverty peers. Researchers should target and test preschool interventions for their potential efficacy to mediate the effects of poverty on early reading."

This study responds to established theory (Chall, 1983; Ehri, 1987, 1995, 1999) that there are discrete stages in reading acquisition and considers the likelihood that young children progress across these stages into conventional reading by the end of first grade.

The study tests the assumptions that developing readers progress through stages of development and the stages have distinct characteristics, are based on solid theory, and they can be measured. This study focuses solely on the foundational knowledge (Letter knowledge, phonemic awareness, and phoneme-grapheme relationship knowledge) needed to read unknown words rather than methods of acquiring sight words.

Research Questions:

1. Given data on important reading indicators for many individuals across the kindergarten and first-grade years, can we form latent classes?

2. Are these classes represented across the reading acquisition window (kindergarten and first grade)?

3. Given membership in a given class, what is the probability of movement into a successive class over time?

4. Is this probability of movement related to poverty status?

Reading achievement was defined by measures of letter recognition, knowledge of initial sounds

in words, knowledge of final sounds in words, word reading in isolation, and reading comprehension (at the sentence level). Reading achievement followed a pattern of mastering low complexity skills prior to mastery of higher complexity skills.

Latent transition analysis allowed researchers to investigate whether there were similar latent classes (discrete phases of development) for groups of individuals in the study at each of four time periods, and to estimate the statistical probability that individual members in a latent class would transition to a more advanced latent class at four different time points from the start of kindergarten to the end of first grade. They found five latent classes existing in different proportions at different times, providing evidence of a sequential order of knowledge: low alphabet knowledge; early phonemic awareness; advanced phonemic awareness; early word reading; early reading comprehension. 

In each class, the relationships among variables were stable. A high probability of passing the beginning sounds task was achieved only after mastery of the alphabet knowledge task. Similarly, the probability of passing the sight word task was achieved only after mastering ending sounds. The most complex skill, identifying words in context, was achieved only after mastery of all other tasks and only during first grade. Early reading comprehension, however, did not appear totally dependent on mastery of less complex tasks for a relatively small number of the sample. 

Poverty Level Analyses

At the start of kindergarten, more children living in poverty level households began reading instruction with low alphabet knowledge compared with their peers living above poverty; by the spring of kindergarten, 75% of children living above poverty had attained at least advanced phonemic awareness, while only 53% of peers living below poverty had. In fall of first grade, 43% of those above poverty were at an early word learning phase or early reading comprehension compared with 21% of those below poverty. By the end of first grade, 87% of the the above poverty group were early word readers, in contrast to 30% of low-SES peers.

Also of interest were transition probabilities for children above and below poverty. Between the fall and spring of kindergarten, children who entered kindergarten with early phonemic awareness were likely to progress, regardless of SES. This sample, however, had large numbers of children living below poverty with low alphabet knowledge, making no movement even following a year of instruction. 60% of the children in the low-alphabet-knowledge latent class above poverty maintained their latent class membership (while 40% moved upward), but 81% of low SES children maintained  latent class membership (while only 19% moved). In the fall of first grade, children above poverty with at least advanced PA progressed beyond that level, while children below poverty, beginning at that same status (advanced PA) were unlikely to make any progress. In fact, the probability of .86 in the below-poverty group failing to advanced by the end of first grade mirrors Juel’s (1988) findings that poor readers, especially those who leave kindergarten with weak phonological skills, experience little reading success later. Children living in poverty who start first grade with advanced PA but not early word learning are much less likely to progress during first grade than their peers living above poverty. 

Author Comments Regarding Intervention: 

"Our study suggests that timing matters, especially for children living in poverty. Between fall and spring of kindergarten and between spring of kindergarten and fall of first grade, children with low alphabet knowledge are unlikely to progress to more advanced understandings. That time, then, is an especially important one for testing the efficacy of interventions and their effect on these transitions. Similarly, all children are unlikely to make progress in their basic reading skills during the summer months between kindergarten and first grade; given this finding and the work of others on the importance of summer (e.g., Alexander et al., 2001), this is another time when the impact of intervention on children’s literacy transitions might be explored. Finally, for those children who enter first grade with advanced phonemic awareness, children living in poverty are much less likely than their peers with more economic advantages to move to early word reading or early reading comprehension; interventions which target the achievement of these children during the first-grade year may have profound effects on their literacy transitions (p. 560)."