5. Results

The results are organized by research question. Because our outcomes variables were raw and not age normed, age was a covariate in our models. For descriptive statistics for child measures by gender and childcare type, see Table 1. For correlations between variables, see Table 2.


Variable

139

1. Child age

2. Gender

3. Center or

0.11

0.00

—

home

4. Marital

0.10

 0.04

0.26\* —

status

5. Mother age 0.23

6. Mother

0.17

0.01 0.10

0.15

0.13

—

workweek

7. Father age 0.23

8. Father

0.11

 0.23

 0.21

0.03

 0.09

0.07

0.02

—

workweek

9. # Siblings

10. Subsidy

11. Mother

0.13

.08

0.50

0.20

 0.35

\*\*

0.13 0.31\*

0.10 0.01

0.48

—

\*\*\*

\*\*\*

education

12. Father

0.13

.02

0.43

0.21

 0.58

\*\*\*

\*\*\*

0.05 0.28\*

0.38\*\*

0.71\*\*

—

0.03 0.53

\*\*\*

education

13. Income

14. TEMA

15. PPVT 16. PALS 17. Verbal

0.27\*

0.09

0.34

0.05

 0.40\*\* 0.03 0.52\*\*\*

\*\*

WM

 0.49

\*\*\*

0.04

0.23\*

0.04

 0.13

0.07 0.13 0.22

 0.55

\*\*\*

0.21

0.21

0.01

 0.20

0.01 0.19 0.02

 0.72

\*\*\*

0.05

0.18

0.04

 0.19

0.14 0.20 0.25

0.01 0.11

0.20

 0.68

\*\*\*

0.51

0.16 0.58

\*\*

\*\*

0.15

 0.41

 0.07 0.01

0.10

0.08

0.06 0.23\*

0.23\*

0.19

 0.29\*

0.06 0.44\*\*\*

0.43\*\*\* 0.42\*\*\*

—

0.08

 0.25\*

0.33\*

0.01 0.73\*\*\*

0.64\*\*\*

—

0.10

 0.38

\*\*

0.41

0.02

 0.70

\*\*\*

—

\*\*

 0.20

 0.36

\*\*

0.00

—

\*\*

0.72\*\*\* 0.45\*\*

0.53\*\* —

0.04

.09 0.15

0.58

0.25

0.16

0.22

0.26

0.08

—

\*\*\*

 0.03

.09

.14

 0.19

 0.42

\*\*

0.01 0.48\*\*

0.14

—

0.02

0.39\*

0.10

 0.87\*\*\*

0.09

—

Predictors of Early Numeracy: Applied Measures in Two Childcare Contexts

DOI: http://dx.doi.org/10.5772/intechopen.81065

 0.02

0.28\*

0.25

—

0.08

—

—

 1

 2

 3

 4

 5

 6

 7

 8

 9

 10

 11

 12

 13

 14

 15

 16

 17

 18

 19

 20

#### Table 1.

Descriptive statistics for child measures by gender and childcare type.

#### Predictors of Early Numeracy: Applied Measures in Two Childcare Contexts DOI: http://dx.doi.org/10.5772/intechopen.81065

First the real words repetition task was presented, then the non-words repetition task. The assessor told the child 'I will say a word and I would like you to repeat it.' If the child had a problem with immature articulation, this was taken into consid-

Three graduate students individually assessed children at the child's out-ofhome care program. Training meetings were held to discuss assessment administration and to recognize test fatigue. The same graduate student administered all assessments with a particular child in two sessions within a one-week time frame. The order of exposure to assessments was randomized based on child preference (in order to maximize cooperation, the child was asked whether they wanted to do

The results are organized by research question. Because our outcomes variables were raw and not age normed, age was a covariate in our models. For descriptive statistics for child measures by gender and childcare type, see Table 1. For correla-

Childcare centers (n = 55) Family childcare (n = 34) Variable Boys (n = 29) Girls (n = 26) Overall Boys (n = 18) Girls (n = 16) Overall

M 100.46 104.31 102.31 98.94 91.19 95.18 SD 13.19 12.58 12.92 17.88 12.29 15.69

M 110.04 104.81 107.52<sup>a</sup> 105.12 98.56 101.94<sup>a</sup> SD 16.53 11.05 20.62 10.09 13.90 12.35

M 66.89 75.31 71.02<sup>a</sup> 63.78 42.56 53.79<sup>a</sup> SD 34.14 34.07 34.04 41.26 35.19 39.44

M 28.61 28.42 28.52<sup>b</sup> 26.17 27.56 26.82b SD 1.52 1.58 1.54 3.94 1.86 3.18

M 32.48 30.35 31.47<sup>b</sup> 42.85 35.75 39.44<sup>b</sup> SD 8.13 5.15 6.91 12.27 6.80 10.50

eration. The reliability for this study was 0.80.

4.2.3 Assessment protocol

Early Childhood Education

numbers or letters first).

TEMA-3 (standard)

PPVT-III (standard)

PALS (total raw)

Means differ at <sup>p</sup> 0.05. <sup>b</sup> Means differ at p 0.01.

a

138

Table 1.

Verbal Working Memory (raw)

BRIEF-P Planning & Organizing/Working Memory (raw)

Descriptive statistics for child measures by gender and childcare type.

tions between variables, see Table 2.

5. Results


\*\*\*p 0.001. Note: Raw scores used.

Table 2. Correlationsbetween

 variables. 5.1 Question 1: Do Children's scores differ by care type on early numeracy, phonological awareness, receptive language, EF, and WM measures?

Predictors of Early Numeracy: Applied Measures in Two Childcare Contexts

3, PPVT-III, and PALS; therefore, parent BRIEF-P scores will not be discussed further. Correlations between phonological working memory and the BRIEF-P working memory subscale were not significant, suggesting the measures assess two

5.2 Question 2: Which of the applied measures (phonological awareness, executive functioning, or phonological working memory) is the best predictor of numerical performance? Does the predictor change when

Since the BRIEF-P Working Memory and Planning and Organizing subscales were highly correlated, the two scales were combined according to the BRIEF-P protocol. For clarity of communication, we will refer to this combination as BRIEF-P Working Memory/Planning & Organizing subscales. Raw scores were used for all

Regression analyses using the enter method were performed to determine which linguistic and working memory measures predicted TEMA-3 performance and to determine if care type was a significant predictor (see Table 3 for regression results). All four models indicated a significant effect: Model 1, including age, care type, BRIEF-P Working Memory/Planning & Organizing subscales, and real words repetition task; Model 2, which substituted non-words repetition task for the real words repetition task; Model 3, which added PPVT-III; and Model 4 which included PALS. The best fit was Model 4 and accounted for 76% of the variance in TEMA-3 performance, with age (t[71] = 4.60, p < 0.001), PPVT-III (t[71] = 2.45, p < 0.05),

We separated our sample into two groups by age (see Table 4 for regression results). The younger group ranged in age from 39 to 55 months (n = 44, M = 48.14, SD = 4.75); the older group ranged in age from 56 to 75 months (n = 44, M = 61.59, SD = 4.00). Regressions were performed to explore differences between outcome variables for the two age groups. Using the enter method we found Model 1 and

separate aspects of memory.

DOI: http://dx.doi.org/10.5772/intechopen.81065

the measures used.

looking at different age groups?

variables and age was corrected within the models.

and PALS (t[71] = 4.95, p < 0.001) as significant predictors.

5.2.1 Predictors of early numeracy performance

5.2.2 Differences between age groups

141

Teacher BRIEF-P, but not parent scores, were significantly related to the TEMA-

A 2 (Gender) X 2 (Caregiving Type) ANOVA was run to determine significant differences in children's scores on all measures. The main effect of gender was significant for PPVT-III standardized scores, F(1, 87) = 3.92, p = 0.05, with boys' scores significantly higher. The main effect of caregiving type was significant for: PALS, F(1, 86) = 5.16, p = 0.03; PPVT-III, F(1, 87) = 4.13, p = 0.05; phonological working memory, F(1, 87) = 10.79, p ≤ 0.001; and BRIEF-P Working Memory/ Planning & Organizing, F(1, 79) = 14.39, p ≤ 0.001, with childcare center scores significantly higher for PALS, PPVT-III, and phonological working memory. BRIEF-P scores were significantly higher for children in family childcare, with higher values indicating more concerns. The interaction (Gender X Caregiving Type) was not significant for any of the measures. No significant differences were found between English-First-Language children and ESL children (n = 3) for any of
