A Comparison of Student Learning Under Multiple Conditions: Classroom instruction, one-on-one human tutoring, and different types of computer tutoring

Neil Heffernan

A major goal for the field of education is to figure out what is the best method to achieve high levels of student achievement. This project addresses this goal by focusing on a single mathematical topic. Using pre & posttests, we assess student's learning gains, as we compare different forms of instruction, over a one hour period of time. At least 3 different forms of instruction will be analyzed: 1) one-on-one human tutoring, 2) computer tutoring, and 3) classroom instruction. The mathematical topic that will be the focus on this work is "algebra symbolization", which translating a real world problem into an algebra expression. Symbolization is hard for algebra students, as NAEP and TIMSS results have shown, nevertheless, the skill is arguably the most important skill students learn in an algebra class. Bloom (1984) reported that one-on-one tutoring is enormously more effective than classroom instruction. One of the goals of this project is to attempt to replicate that result as well as compare several varieties of computer based tutoring. Another goal is to learn techniques classroom teachers can learn from effective human tutors. Bloom reported these results with a very small sample size in terms of the number of teachers and students. This project aims to get over one hundred teachers participating. This project, which started as an individual research project by Neil Heffernan, has expanded into a multi-disciplinary collaboration between different research groups and classroom teachers. The project is now called that Learning Open (www.LearningOpen.org). Classroom teachers and other researchers are invited to participate by using the same pre & posttest to assess student learning.

About Neil Heffernan

Dr. Neil Heffernan graduated summa cum laude from Amherst College in History and Computer Science. Neil taught mathematics to eight grade students in Baltimore City as part of Teach for America, a program that selectively recruits top candidates to teach in inner-city schools. After observing the sorry state of educational software, Neil went off to Carnegie Mellon University to do research in creating educational software that leads to higher student achievement. For his dissertation, he built the first intelligent tutoring system that incorporated a model of tutorial dialog. This system has been shown to lead to higher student learning, by getting students to think more deeply about problems. It is based upon detailed studies of student learning as well as studies of experienced human teachers. The system (free at www.AlgerbaTutor.org) has been used by thousands of students and teachers and has been awarded many educational awards. Neil is now a assistant professor at Worcester Polytechnic Institute, where one of his projects is organizing "The Learning Open" (www.LearningOpen.org), an interdisciplinary collaboration between educational software researchers and classroom teachers to study what are the benefits of different instructional approaches. While on holiday, Neil and his wife (who is also a math teacher and researcher) talk to their one-year old son about mathematics.