Theses Doctoral

Is Problem Based Learning Effective in Fostering the Development of Intellectual Skills?

Jewett, Elizabeth

How best to help students master the control of variables skill is a debated topic among educators and psychologists. Contrasting theories as to how it develops parallel contrasting ideas about how it should be taught. Recent studies relying on direct instruction have shown that this approach to teaching the control of variables skill is not successful for academically disadvantaged populations.
Problem Based Learning (PBL) is a method of instruction that holds promise in this regard and is explored in the present work as a method for teaching the control of variable skill. Recent empirical studies have proven its effectiveness with the learning and application of declarative knowledge. A tightly controlled experiment was conducted with 6th and 7th grade students. Students were randomly assigned to one of 4 conditions (team, individual, observer, control) following a pretest to determine skill level. These groups varied in level of engagement and social interaction. Students who solved the problem individually, in a team of three, or students that observed an individual solver were compared to a control group. Students were presented with a realistic problem involving causes of teen crime. Records from varying towns reported variable levels and outcomes. Over three sessions students examined the records with the objective of determining which variables did or did not make a difference to teen crime. Following the PBL sessions students in all conditions were administered both immediate and delayed assessments of skill..
Between 75 and 80% of students in the PBL individual and team conditions successfully demonstrated use of the skill. Students who achieved the highest level on both the causal and noncausal variables scored on average 41% higher on the posttest. There was no difference between these two groups in acquisition of the skill. A significant difference was found between the the PBL groups and the two remaining groups (observer and control). Performance of these groups was not significantly different and inferior to that of the PBL groups..
The implications of these findings are considered in a number of different theoretical and applied contexts: a) enhancing students' mastery of the control of variables skill; (b) extending evidence of the effectiveness of a PBL method from declarative to procedural knowledge; (c) active vs. vicarious learning, and (d) individual vs. collaborative reasoning and problem solving.


  • thumnail for Jewett_columbia_0054D_12123.pdf Jewett_columbia_0054D_12123.pdf application/pdf 2.51 MB Download File

More About This Work

Academic Units
Thesis Advisors
Kuhn, Deanna
Ph.D., Columbia University
Published Here
July 7, 2014