Panel Paper: Kindergarten Mathematics Instruction and the Common Core

Thursday, November 6, 2014 : 11:15 AM
Enchantment II (Convention Center)

*Names in bold indicate Presenter

Tyler Watts, George Farkas and Greg J. Duncan, University of California, Irvine
The current study investigates whether kindergarten mathematics instruction has changed during the past decade, as standards-era reform movements have attempted to overhaul K-12 education.  We conceptualize mathematics instruction along Common Core dimensions, and investigate whether changes in time spent on Common Core topics relate to student achievement gains during the kindergarten year.

Data come from the both the 1998-1999 and 2010-2011 samples of the Early Childhood Longitudinal Study- Kindergarten Cohort datasets that contain independent, nationally representative samples of children recruited at the beginning of kindergarten. We used teacher report measures in which teachers indicated how much time per month was spent on various mathematics instructional topics (e.g. counting to 10, adding and subtracting single-digit numbers).  We then categorized instructional topics into the domains specified by the Kindergarten level of the Common Core Standards for Mathematical Practice (CCSSM). This categorization led to three broad categories (Kindergarten CCSSM Topics; Advanced CCSSM Topics (topics beyond kindergarten); Non-CCSSM Topics). 

The samples (1999: n= 15,090; 2010: n= 12,880) were restricted to students who: 1) could be matched to the same kindergarten teacher in fall and spring; 2) had valid fall and spring test scores; 3) had instructional data from the spring. We used a dummy variable approach to handle missing data on covariates.

We found that kindergarten teachers report spending significantly more time on mathematics instruction in 2010-2011 than in 1998-1999 (see Table 2).  The increase in time spent on mathematics instruction is shared amongst Kindergarten CCSSM Topics, Advanced CCSSM Topics, and Non-CCSSM Topics. Further, we did not find that children receive different amounts of time on mathematics instruction based on their school-entry test scores, but children in full-day kindergarten receive significantly more mathematics instruction than children in half-day kindergarten (see Table 3).   We did not find that socioeconomic status or race was related in any clear way to instructional time.

  Finally, we find that time spent on certain mathematical domains recommended by the CCSSM (Operations and Algebraic Thinking; Number and Operations in Base Ten) is more strongly associated with kindergarten achievement gains than time spent on other domains (Geometry; Measurement and Data) in both the 1998 and 2010 cohorts (see Table 4).  However, time spent on advanced CCSSM domains appears to have a stronger relationship with achievement gains in the 2010 cohort than the 1998 cohort.

We investigated whether time spent on various instructional topics differed based on certain child characteristics.  In both samples, interactions between levels of the fall math score (low, average, high) and instructional topic failed to produce significant patterns, and joint-tests revealed that the set of interaction variables did not significantly contribute to the models.  This same pattern held for interactions between family socioeconomic status and time spent on instructional topic.  These results lend further support to the findings reported by Claessens, Engel and Curran (2013), as the benefits of advanced mathematics instruction during early childhood appear to be shared across students at all levels of school-entry skills and socioeconomic status.