Skip Navigation
A Study of Classroom Literacy Interventions and Outcomes in Even Start

NCEE 2008-4028
September 2008

Impact Findings

Impacts of the CLIO Combined Curricula. The study showed that Even Start projects assigned to the CLIO combined curricula did not exhibit better child language and literacy outcomes than Even Start projects assigned to the control group (figure ES-1). In the figures in this section, effect sizes for the combined curricula are indicated by filled diamonds (relative to the control group) and open circles (relative to the preschool curricula), and 95 percent confidence intervals5 are shown as horizontal bands on either side of the diamond or circle. Effect size indicates the difference in outcome between the average subject who received the treatment and the average subject who did not.6

There were no statistically significant impacts of the CLIO combined curricula on any of the seven measures of child language and literacy skills (six in English and one in Spanish), as can be seen by the fact that none of the confidence bands exclude zero, even before adjustment for multiple comparisons. Estimated effect sizes on emergent literacy outcomes were all smaller than 0.13 in absolute value, with confidence interval limits all bounded by 0.27 in absolute value. However, the CLIO combined curricula did have a statistically significant positive effect on child social competence (behavior in class) as rated by preschool teachers. The effect size of the impact of the CLIO combined curricula on child social competence was 0.22.

The CLIO combined curricula had a statistically significant positive impact on both of the parent outcomes examined (figure ES-2). The effect size of the impact on parent interactive reading skill was 0.48, and the effect size of the impact on parent responsiveness to their child was 0.22. Even though CLIO did not manipulate adult education curricula, the study assessed parent reading skills and vocabulary and showed that the CLIO combined curricula did not have a statistically significant impact on these skills (figure ES-2).

The CLIO combined curricula had a statistically significant positive impact on two of five measures of instructional support for literacy development in preschool classrooms (figure ES-3). The effect sizes of the statistically significant impacts on support for print knowledge and literacy resources in the classroom were 0.69 and 0.52, respectively. There was no statistically significant impact on the following three preschool instructional measures: support for oral language development, support for phonological awareness, or support for print motivation.7

The CLIO combined curricula had a positive impact on one of the three measures of parenting education and parent‐child classroom instruction (figure ES-3). The effect size of the impact on the amount of parenting education time spent on child literacy was 1.01. There was no statistically significant impact on the amount of parenting education time spent on parenting skills not related to child literacy or the amount of parent‐child time spent with parents and their children interacting on child literacy activities.

The study also examined whether the CLIO combined curricula had an impact on participation levels (figure ES-4). The results showed that there was no statistically significant impact of the CLIO combined curricula on either child levels of participation in preschool or parent levels of participation in parenting education or parent‐child activities. Neither of the confidence bands exclude zero.

Added Value of the CLIO Parenting Curricula. CLIO parenting curricula did not add significantly to the effectiveness of the CLIO preschool curricula on any of the seven measures of child literacy skills or on child social competence (figure ES-1). That is, adding research‐based parenting components focused on child literacy did not add significantly to childrenʹs outcomes beyond what was achieved with the CLIO preschool curricula. (In figures ES-1 through ES-4, the effect sizes for the added value of the CLIO parenting curricula are indicated by open circles.) The estimated effect sizes of the CLIO parenting curricula on emergent literacy outcomes were all smaller than 0.11 in absolute value, with confidence interval limits all bounded by 0.23 in absolute value.

However, the CLIO parenting curricula did have a statistically significant positive incremental effect on parent interactive reading skill (effect size of 0.30) (figure ES-2). The difference on parents’ responsiveness to their child between the CLIO combined curricula and the CLIO preschool curricula, while similar in size to the statistically significant difference between the CLIO combined curricula and the control group, was not statistically significant.

There were statistically significant incremental effects of the CLIO parenting curricula on two of the instructional measures. The effect sizes of the incremental effects of the CLIO parenting curricula on the amount of parenting education time spent on child literacy and the amount of parenting education time spent on parenting skills not related to child literacy were 0.68 and ‐0.45, respectively (figure ES-3). There was no statistically significant incremental effect of the CLIO parenting curricula on how time was spent in parent‐child classes or (as expected) in preschool classes.

Finally, the CLIO parenting curricula did not have a statistically significant incremental effect on child participation in preschool or on parent participation in parenting education (figure ES-4).

Top


5 The confidence intervals may be interpreted as follows. If the experiment were to be independently repeated a very large number of times under the same general conditions, drawing on the same population of schools and students, and on every repetition both an effect estimate and a confidence interval on that estimate were calculated, then, over the long run, 95 percent of the confidence intervals would contain the long‐run average of estimated effects.
6 Effect size was calculated by taking the difference between the treatment and control group means and dividing that difference by the standard deviation of the control group’s scores in 2005.
7
Although the confidence bands for support for phonological awareness and support for print motivation exclude zero, the effect sizes are not significant once adjusted for multiple comparisons.