Kurt VanLehn, Professor, School of Computing & Infomatics, Arizona State University

 

Kurt VanLehn, Professor, School of Computing & Infomatics, Arizona State University

Sage 4101

November 6, 2013 12:00 PM - 1:30 PM

 

 

Abstract:  It is often said that human tutors are 2 standard deviations more effective than classroom instruction and the best tutoring systems are 1 standard deviation more effective.  This hypothesis, which inspired many important studies of human tutoring and many efforts to replicate human tutoring with natural language tutoring systems, now seems false.  Although research continues, the current best-fitting hypothesis is that both human tutors and the best types of tutoring systems have roughly the same effect size, namely 0.75 standard deviations above no-tutoring instruction.   Part of the talk will support this claim with a meta-analysis of relevant experiments, illustrated with specific experiments from several labs.  However, this finding does not imply that tutoring system developers should declare victory and retire.  The studies found that both human tutors and tutoring systems were far from perfect.  The second part of the talk describes a particularly effective class of tutoring systems, called step-based tutoring systems, that are moderately easy to construct initially and can be semi-automatically improved using big data.  

 

The Relative Effectiveness of Human Tutoring, Intelligent Tutoring Systems, and Other Tutoring Systems

 

 

Add to calendar
Share|