Overall Reading and Mathematics Knowledge and Skills (Scale Scores). One way to analyze differences in children's reading and mathematics knowledge and skills by child, family, and school characteristics is to look at their overall reading and mathematics scale scores. In this report, a standardized representation of the scale score (i.e., t-score) is used. T-scores provide a norm-referenced measure of performance that can be used as an indicator of the extent to which a specific group of children (e.g., males versus females) ranks higher or lower than the national average and how this relative ranking changes over time. The ECLS-K reading and mathematics t-scores are designed to have a mean of 50 and a standard deviation of 10 (when all scores from the complete sample of children are in the analysis). In this report, however, the spring of kindergarten and the spring of first grade mean t-scores are 51 (tables 3 and 4). This is because the analyses in this report are limited to children in the ECLS-K who entered kindergarten for the first time in the fall of 1998, were promoted on time to first grade, and were assessed in English at all three points in time.

In reading and mathematics, differences in children's achievement by their family's poverty status, their race/ethnicity, and their school type persist from kindergarten through the spring of first grade. However, children's reading and mathematics achievement does not vary by their sex (tables 3 and 4).

In both reading and mathematics, children who are poor (i.e., come from families with incomes below the poverty threshold¹³) consistently score about a half-standard deviation below the national average (i.e., 5 to 7 points). At all three points in time, their achievement is significantly lower than that of nonpoor children (tables 3 and 4).

Racial/ethnic differences also exist in children's kindergarten and first grade reading and mathematics achievement. In both reading and mathematics, White children typically score near the national average (i.e., 51). The one exception is in fall kindergarten math. White children enter kindergarten with a mean score about 3 points above the national average. The picture for Black children is slightly different from that of White children. In reading, Black children enter kindergarten 3 points below the national average, and remain 3 points below the national average through the spring of kindergarten and the spring of first grade. In math, Black children are also significantly below the national average at all three points in time (i.e., 3 to 4 points) (tables 3 and 4).

Hispanic children's reading achievement seems to be moving upward toward the mean.¹⁴ When Hispanic children enter kindergarten they score significantly lower than the national average in reading (4 points), but by the spring of first grade they are on par with the national average (i.e., within 1 point). In terms of mathematics, Hispanic children score near the national average (i.e., within 2-3 points) in the spring of kindergarten and the spring of first grade. In reading, Asian children consistently are about a half-standard deviation above the national average (3 to 5 points). In mathematics, Asian children are above the national average during their kindergarten year; but by the spring of first grade, they are within 2 points of the national average (tables 3 and 4).

A slightly different way to think about racial/ethnic differences is to compare the groups to one another, versus comparing the groups to the national average. When the achievement of the different racial/ethnic groups is compared, White and Asian children score significantly higher than Black children in reading and mathematics at all three points in time (tables 3 and 4).

Children's reading and mathematics achievement also differs by school type during their kindergarten year. Children who attended private schools during the kindergarten year enter kindergarten scoring significantly higher in reading and mathematics than the national average, and maintain this difference through the spring of first grade (tables 3 and 4).

13 Poverty is a function of household income and household size. For more information, see the Methodology and Technical Notes section.
14 The children included in the estimates in this report are children who were assessed in English at all three points in time. To explore how including children who demonstrated sufficient English language proficiency to receive the English battery in the spring of kindergarten and spring of first grade (i.e., screened into the battery) would impact the estimates, an example analysis using the reading t-scores was run. Comparing the mean t-scores presented in table 2 for children assessed in English at all three points to the mean t-scores for all children who were screened into the English assessment over time, did not yield significant differences. Specifically, the estimates are as follows: mean t-score fall kindergarten, spring kindergarten, spring first grade for children assessed in English at all points in time-51, 51, 51, respectively; mean t-score fall kindergarten, spring kindergarten, spring first grade for children who screened in over time-51, 51, 51, respectively. Further, the same nonsignificant finding is true in terms of specific racial/ethnic groups (e.g., the estimates are as follows: mean t-score fall kindergarten, spring kindergarten, spring first grade for Hispanic children assessed in English at all points in time-47, 49, 50, respectively; mean t-score fall kindergarten, spring kindergarten, spring first grade for Hispanic children who screened in over time-47, 48, 49, respectively. For more information (e.g., methods and breakdowns of children not assessed in English), refer to the Methodology and Technical Notes section.

<< back    >> next