Tutoring Online for Problem-Based Learning: Effects of Different Tutoring Strategies and Learners’ Expertise

• In Event: Does Feedback in Problem-Based Learning Enhance Student Learning?

Abstract

1. Objectives
Within Problem-Based Learning (PBL; e.g., Barrows, 1985) the tutoring process is a crucial element of the learning process (e.g., Dolmans et al., 1994). While current research provides no clear evidence, when a tutor should be an expert in the field of the problem or not, the tutor’s role is crucial to guide small group discussion. Second, the role of learners’ level of expertise is relevant. Learners with little or no prior knowledge might benefit more from tutor-guided learning than from primarily self-directed learning. This study has been conducted in order to examine how different tutoring strategies might interfere with different learner characteristics under the special conditions of online group tutoring.

2. Theoretical Framework
The basic background of this study is founded in research as well as theoretical models on tutoring. Especially the work by Chi et al (2001) provides here a framework that shows that tutoring might have different objectives which also might interact with several accompanying factors within a learning environment. The question here is how polarized behavior regarding tutor-centered, student centered and interactive coordination focuses (cf. Chi et al., 2010) might have impact on learners with different levels of expertise.

3. Method
Based on prior studies, we examine here in a controlled experiment the role of tutors’ content expertise during a Problem-Based Learning scenario in the area of Clinical Psychology that is performed online by means of synchronous computer-mediated communication.

4. Data Sources and Materials
In a first condition expert tutors acted as a moderators providing no information regarding the problem. In a second condition, the tutors provided learners with his expertise and corrected wrong statements or answered direct questions. A second quasi-experimental was level of students’ expertise (graduated psychologists or other professionals within the area of Clinical Psychology vs. first year students).

5. Results and Conclusions
An overall MANOVA reveals overall effects of both independent variables on measure of knowledge acquisition and knowledge self-confidence as dependent measures but no interaction effects. Between-subjects-effects show that learners with high prior knowledge perform better in a knowledge post-test than learners with low prior knowledge (p=0.01; η2=0.08), independent from expert or non-expert tutoring. In addition, learners benefited regarding knowledge acquisition more from expert tutoring (p=0.03; η2=0.06). Regarding measures of participants’ self-confidence about their knowledge in the area of the problem results reveal main effects for both independent variables. Learners with high prior knowledge (p<0.05; η2=0.05) as well as learners in the non-expert tutoring conditions (p<0.05; η2=0.04) reported to be more self-confident in their abilities.

6. Scientific Significance
Results suggest that learners benefit here in this study more when the tutor imparted his/her knowledge. In addition, the role model an expert tutor is able to provide helps to calibrate learners own progress more accurate and, thus, contributes to avoid an “illusion-of-knowledge”-effect. Finally, findings support that novice learners benefit more from a higher level of instruction while this effect fades with students’ growing expertise. Further research has to examine the “illusion-of-knowledge”-effect in feedback research more deeply.

Authors

• Joerg Zumbach, University of Salzburg

• Jasmin Ottitsch, University of Salzburg