Skip Navigation
Effectiveness of Reading and Mathematics Software Products

NCEE 2009-4041
February 2009

First-Year Effects of First Grade Technology Products

  • Effects on Test Scores Were Not Statistically Different from Zero. Overall reading scores for students in treatment and control classrooms were 50.2 and 49.5, respectively (in normal curve equivalent units).1 The difference was not statistically different from zero.
  • Most School and Classroom Characteristics Were Uncorrelated with Effects. Classroom characteristics (teaching experience, teacher gender, teacher education level, whether there were problems getting access to the product, whether teachers had adequate time to prepare to use the product, whether the product was used in the classroom, and whether the teacher participated in technology professional development in the past year) were not correlated with product effects for the overall SAT-9 score. School characteristics (percentage of students eligible for free lunch, whether the school is in an urban area, percentage of students that were African American, percentage that were Hispanic) also were not correlated with product effects on the overall SAT-9 score. The one exception was the student-teacher ratio. Time of study product usage did not have a statistically significant correlation with effects for the overall score or subtest scores.
1 A normal curve equivalent (NCE) score converts the scaled test score into the range 1 to 99, with 50 being the average for the nationally normed sample. Unlike percentiles, NCE scores can be averaged, which makes them more appropriate for statistical analyses and estimation of product effects.