Poster #72 - Development of Reading and Mathematics Skills in Early Grades: A Parallel Process Latent Growth Model

• In Event: 2-110 - Poster Session 07
  In Poster Session: PS 07 Section - Education, Schooling

Integrative Statement

Purpose of the Study
A growing body of research has indicated that the development of reading and mathematics skills is highly related in early grades (e.g., Claessens et al., 2009; Michalczyk et al., 2013). However, there are only a few studies examining young students’ reading and mathematics development at a national level. While understanding students’ early academic development within the local context can provide useful information to teachers and school administrators, examining its development patterns and interrelationships at the national level can enhance policy discussion and better inform policymakers and stakeholders. This study contributes to the literature by examining the concurrent development of reading and mathematics skills from kindergarten to third grade, and their relations using data collected from a nationally representative sample.
Data and Analytic Methods
This study uses data from the Early Childhood Longitudinal Study Kindergarten Class of 2010-2011, which was administrated to a nationally representative sample of kindergarten students. To examine early academic development, this study uses seven measurement occasions: fall kindergarten, spring kindergarten, fall first grade, spring first grade, fall second grade, spring second grade, and spring third grade. At each time point, direct cognitive assessments were individually administered for reading and mathematics, and these assessments were vertically scaled, allowing longitudinal analysis. The reading assessment measures basic literacy skills, such as phonemic awareness, vocabulary, and passage comprehension. The mathematics assessment measures knowledge and problem-solving skills related to number and operations, measurement, probability, and so on. Student demographic characteristics (e.g., gender, race/ethnicity, and family SES) will be included for final analysis.
In the parallel process latent growth model (LGM), growth of repeated measures of two outcomes (i.e., reading and mathematics scale scores) are simultaneously modeled, assuming the two growth trajectories are associated with each other. The unconditional parallel LGM with seven time point repeated measures is shown in the path diagram (see Figure 1). The covariance between the latent growth factors (intercept and slopes) of the two growth trajectories provides information about the association between the initial status and growth rates in students’ reading and mathematics skills. As both reading and mathematics skills change over time in a nonlinear way, we use quadratic instead of linear LGM to fit the data.
Results and Significance
Our preliminary analysis shows that the initial levels of reading and mathematics skills were highly positively correlated with each other (r=0.85, p<0.01). The linear slopes were also moderately positively correlated (r=0.36, p<0.01), suggesting that those with faster growth in reading also had faster growth in math. For reading, initial skills in fall kindergarten are negatively related to their growth (r=-0.43, p<0.01), while initial skills in mathematics are positively related to mathematics growth (r=0.27, p<0.01). Additionally, reading skills in fall kindergarten are positively related to mathematics growth (r=0.26, p<0.01), whereas mathematics skills in fall kindergarten are negatively related to reading growth (r=-0.31, p<0.01).
Preliminary analysis of unconditional parallel LGM suggests while students’ performance in reading and mathematics are highly related, their growth trajectories in one subject may not contribute much diagnostic information for another.

Authors

• Mengyi Li, American Institutes for Research
  Presenting Author

• B. Jasmine Park, American Institutes for Research
  Non-Presenting Author

• Yuan Zhang, American Institutes for Research
  Non-Presenting Author

• Thomas Sun, American Institutes for Research
  Non-Presenting Author

• Xiaying Zheng, American Institutes for Research
  Non-Presenting Author