School Readiness Early Literacy Trends for Preschoolers ...



School Readiness Early Literacy Trends for Preschoolers Entering Kindergarten

A Research Paper Presented

By

Kevin McGowan

To

Dr. Dimiter Dimitrov

Instructor

In partial fulfillment of the requirements for

Quantitative Methods in Education Research

George Mason University

Fairfax, Virginia

June 18, 2008

Introduction

Research regarding early childhood classrooms has provided evidence for the positive impact of high quality preschool classrooms on children’s early language and literacy skills (Dickinson & Caswell, 2007). Studies that have examined specific features of preschool classrooms have found relationships between classroom quality and children’s language and early literacy skills (Dickinson & Caswell, 2007). The Head Start Family and Child Experiences Survey (FACES) provide evidence that preschool children attending Head Start can verbally express their full name and age, identify ten basic colors by name, use a pencil to write letters of the alphabet, especially those letters that appear in their names, and answer simple questions about a story that was read to them.

Leaper and Smith (2004) examined three sets of meta-analyses for gender effects on preschool children’s language use. On average, girls were more talkative than boys; whereas, boys used more assertive speech than girls (Leaper & Smith, 2004). The below basic literacy skills for African-American children attending urban schools is well established. Intervention efforts have traditionally focused on kindergarten through fifth grade; however current research suggests that intervention efforts start prior to kindergarten (Washington, 2001). Latino children attending urban schools do not perform any better than their African-American counterparts regarding basic literacy skills; therefore, early literacy intervention efforts should also target inner-city Latino children. This study is intended to examine the effects of gender and ethnicity on early literacy in preschool children attending center-based programs via the following research questions.

1. Are there differences in pretest letter naming assessment scores according to gender?

2. Are there differences in pretest letter naming assessment scores according to ethnicity?

3. Do pretest letter naming scores predict posttest letter naming scores?

4. Can posttest letter naming assessment scores be predicted from pretest letter naming assessment scores and from pretest name writing scores?

Method

Sample

The sample consists of 803 preschool children with letter recognition pretest data and 853 preschool children with letter recognition posttest data from the National Child Care Research Database. The children are from low, middle, and upper income families. The sample is 49% male and 51% female. The racial/ethnic composition of the sample is 46% White, 27% African American, 15% Latino, 9% Multiracial, 2% Asian American, and 1% Native American.

Data

The letter naming outcomes are assessed via teacher generated assessment tools, the Houghton-Mifflin web-based pre-kindergarten assessment system, work sampling online, the pre-kindergarten version of the Phonological Awareness Literacy Screening (PALS), and Head Start National Reporting System (NRS) data. The pretest was administered in the fall of the academic school year and the posttest was administered in the spring of the academic year. Ninety percent of the assessments were administered by the child’s teacher and 10% were administered by early childhood education specialists. Data Collection

Data was collected from the National Child Care Research Database. The database includes assessment scores, the number of boys and girls, ethnicity, and family socio-economic status. Data was collected in accordance with Human Subjects Review Board criteria. Informed consent forms were collected from all parent participants and informed assent forms were collected from student participants.

Statistical Data Analysis

An independent samples t-test was done to determine if there is a mean difference between boys and girls on pretest letter naming assessment scores. A one-factor analysis of variance was done to determine if there is a mean difference between ethnic groups on pretest letter naming assessment scores. A post hoc Tukey test was done to check for specific differences between ethnic groups. A simple linear regression was done to determine if pretest letter naming assessment scores are a good predictor for posttest letter naming scores. A multiple regression was done to determine if pretest letter naming assessment scores and pretest name writing assessment scores are good predictors for posttest letter naming scores.

Results

The alpha level for all tests is .05.

Research Question One

Are there differences in pretest assessment scores according to gender? Results from an independent samples test show that the p-value in the Levene test for equality is .003 which is less than .05; therefore, the null hypothesis is rejected. The t-test for equality of means test shows that there is a statistically significant difference between females and males on pretest letter knowledge, t(800) = 2.07, p = .003. The 95% CI = (.07, 2.42). Thus, females could name more letters of the alphabet by at least 1, but no more than 2.

Research Question Two

Are there differences in pretest letter naming assessment scores according to ethnicity? Results from a one-way analysis of variance show that the p-value is .001 which is less than .05; therefore, the null hypothesis is rejected. There are statistically significant differences among the six ethnic groups (Latino, African American, Native American, Asian American, White, and Multiracial), F(5, 789 = 4.23), p = .001. Further, the Post Hoc Tukey results show that there are statistically significant differences between Latino preschoolers and Asian American preschoolers (p = .003), between African American preschoolers and Asian American preschoolers (p = .016), and between White preschoolers and Asian American preschoolers (p = .002). Specifically, the 95% CI shows that: (a) Asian American preschoolers could name at least 2 more letters of the alphabet but not more than 17 in relationship to Latino preschoolers; (b) Asian American preschoolers could name at least 1 more letter of the alphabet but not more than 15 in relationship to African-American preschoolers; and (c) Asian-American preschoolers could name at least 2 more letters of the alphabet but not more than 17 in relationship to White preschoolers.

Research Question Three

Do pretest letter naming scores predict posttest letter naming scores? Results from a simple linear regression show that the p-value < .05 which means that the results are statistically significant. The correlation coefficient (R = .800) is statistically significant at the .05 level, F(1, 744) = 1320.412, p = .000. The coefficient of determination is R² = .640, thus indicating that 64% of the variance in the posttest scores is accounted for by the variance in the pretest scores. The simple linear regression equation for the prediction of posttest scores from pretest scores is: Ŷ = (0.886) pretest scores + 5.823.

Research Question Four

Can posttest letter naming assessment scores be predicted from pretest letter naming assessment scores and from pretest name writing scores? The F-statistic is statistically significant, F(2, 733) = 708.942, p = .000, thus providing evidence that the variance in posttest letter naming assessment scores accounted for by the two predictors does not equal zero for the population. Specifically, the coefficient of determination R² = .659 indicates that 65.9% of the preschoolers’ differences in posttest letter naming assessment scores are accounted for by their differences in pretest letter naming assessment scores and pretest name writing assessment scores. The multiple regression equation for predicting posttest letter naming assessment scores from pretest letter naming assessment scores and pretest name writing assessment scores is: Ŷ = 0.790(pretest letter naming assessment scores) + 0.039(pretest name writing scores) + 4.396. In this case, each independent variable is statistically significant at the .05 level (p = .000 for both variables. The part correlation between posttest letter naming assessment scores and pretest letter naming assessment scores, partialling out pretest name writing assessment scores, is rү(1.2) = .591. Thus (.591)² = .3493 shows that 34.93% of the variance in posttest letter naming assessment scores is uniquely accounted for by the variance in pretest letter naming assessment scores. Likewise, the part correlation between posttest letter naming assessment scores and pretest name writing assessment scores, partialling out pretest letter naming assessment scores from pretest name writing assessment scores, is rү(2.1) = .129. Thus (.129)² = .0166 shows that 1.66% of the variance in posttest letter naming assessment scores is uniquely accounted for by the variance in pretest name writing assessment scores. These results indicate that pretest letter naming assessment scores is more important than pretest name writing assessment scores for the explanation or prediction of posttest letter naming assessment scores.

Discussion

This study provides some conclusions related to gender and race of preschoolers and their ability to name letters of the alphabet. The data indicated that there were differences between females and males in terms of naming letters with females outperforming males. This result is supported by the research done by Leaper and Smith (2004).

In terms of ethnicity, Asian-American preschoolers were able to name more letters than their Latino, African American, and White counterparts. There were no statistically significant differences observed between White preschoolers and African-American preschoolers or between Latino preschoolers and African-American preschoolers. In general, pretest letter naming scores are an accurate predictor of posttest letter naming scores as evidenced by linear and multiple regressions.

Some limitations to consider regarding this study are: posttest letter naming scores were not examined for gender and ethnic differences, the socio-economic status of the children was not examined, and the number of years of center-based child care was not examined. In addition, the small number of Asian-American preschoolers compared to Whites, African Americans, and Latinos may have skewed final results.

Future studies can further explore the impact of socio-economic status on preschoolers’ ability to name letters. Additional research can examine the number of years that a child has spent in center-based child care facilities and its impact on the child’s ability to name letters.

References

Administration for Children, Youth, and Families. (2001). Head Start FACES: Longitudinal findings on program performance. Third progress report. Washington, DC: Department of Health and Human Services. _title.html.

Dickinson, D. K. & Caswell, L. (2007). Building support for language and early literacy in preschool classrooms through in-service professional development: effects of the literacy environment enrichment program. Early Childhood Research Quarterly, 22, 243-260.

Leaper, C. & Smith T. E. (2004). A meta-analytic review of gender variations in children’s language use: talkativeness, affiliative speech, and assertive speech. Developmental Psychology, 40, 993-1027.

Washington, J. A. (2001). Early literacy skills in african-american children: research considerations. Disabilities Research and Practice, 16, 213-221.

Table 1

Summary of One-Factor Analysis of Variance

________________________________________________________________________Source df F ή p

________________________________________________________________________Ethnicity 5 4.23 .16 .001

S within-group

error 789 (71.52)

Note. Values enclosed in parentheses represent mean square errors. S = subjects.

Table 2

Summary of Linear Regression Analysis

________________________________________________________________________

Independent Variable df Mean Square F p-value

Pretest 1 42987.689 1320.412 .000

________________________________________________________________________

B SE B β p-value

______________________________________________________

.886 .024 .800 .000

a. Predictors (Constant), Pretest Letter Naming Assessment Scores (%)

b. R = .800; R² = .640.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download