Predicting Fourth Grade Student English Language Art (ELA) Achievement Among New York Schools

Abstract

The researchers used structural equation modeling (SEM) to create a model to predict fourth-grade student achievement in English Language Arts (ELA) by exploring the relationships among: student, household, school, and teacher factors. Public data from the New York State Education Department (NYSED) 2012–13 school report card data, NYSED fiscal reporting system, Census 2010 School District Demographics System, and 2011 Civil Right Data Collection were used from 1,263 schools in New York excluding New York City. Variables were chosen using this convenient sample and supported by our conceptual rationale.

The model predicted fourth-grade ELA achievement with 73 percent of effect size. Household attributes had the strongest predictive value for student achievement. School attendance rate, and the percentage of teachers with master’s degrees plus 30 additional educational credits in the school had medium predictive value for student achievement.

The model also predicted the percentage of economically disadvantaged (ED) students in a grade with 73 percent of effect size based on the school district household attributes. Single family household was a major predictor for ED students.

Keywords: Path Analysis, SEM, Achievement, Household, SES, Attendance