Searching for the two sigma advantage: Evaluating algebra intelligent tutors

• We evaluated two intelligent tutoring systems in a summer school program.
• We examined test scores among repeated measures of arithmetic and algebra.
• Test scores increased over time and produced large pre/post effect sizes.
• Test scores between the two systems were not significantly different.

This study evaluated 2 off-the-shelf, computer-based, mathematics intelligent-tutoring systems that provide instruction in algebra during a remedial mathematics summer program. The majority of the enrolled high school students failed to pass algebra in the previous semester. Students were randomly assigned in approximately equal proportions to work with the Carnegie Learning Algebra Cognitive Tutor or the ALEKS Algebra Course. Using the tutoring system exclusively, the students completed a 4-h-a-day, 14-day summer school high school algebra class for credit. The results revealed that both tutoring systems produced statistically and practically meaningful learning gains on measures of arithmetic and algebra knowledge.