2018 ESA Annual Meeting (August 5 -- 10)

COS 82-8 - Learning via inquiry and feedback: trade-offs between open-endedness and constraint in simulated ecology and evolution labs

Wednesday, August 8, 2018: 4:00 PM
245, New Orleans Ernest N. Morial Convention Center
Eli Meir1, Denise Pope2, Kerry Kim1, Daniel Wendel3, Ling Hsiao3, Susan Maruca1, Jody Clarke-Midura4 and Joel K. Abraham5, (1)SimBio, Missoula, MT, (2)CIRTL, Northampton, MA, (3)Teacher Education Program, Massachusetts Institute of Technology, Cambridge, MA, (4)Instructional Technology and Learning Sciences, Utah State University, Logan, UT, (5)Biological Sciences, California State University, Fullerton, Fullerton, CA
Background/Question/Methods

Inquiry-based teaching tools can take many forms. Inquiry-based learning environments range from highly constrained, where students follow a pre-set path through an activity and answer a series of questions, to highly open-ended, where students pose and answer their own questions with minimal constraints. Previous research has suggested that learning environments at both extremes are often less effective than those with intermediate levels of constraint. For educational software to provide instant feedback, some constraints are generally required.

We examined how level of constraint and the corresponding ability to provide feedback impacted how students engaged with inquiry-based educational software. Here we synthesize a large amount of student data from several simulation-based ecology and evolution labs representing a range of constraints on simulation activities and embedded questions. Question types in these labs include multiple-choice (highly constrained), LabLibs and WordBytes (formats found in some SimBio Virtual Labs® that are moderately constrained), and short essay (open-ended). We investigate usability of formats with varying constraint levels, the ability of students to express higher-order thinking, student engagement, how student answers compare, and generally how constraint and feedback promote learning.

Results/Conclusions

This talk integrates data from several studies involving over 7,000 students at over 45 high schools and colleges using five different simulation-based ecology and evolution activities. We examined usage patterns for several versions of SimBio’s lab “Understanding Experimental Design” with three levels of constraint and feedback specificity built in. We also examined usage patterns for 17 questions in two intermediate constraint formats as compared to 4 short answer and dozens of multiple-choice questions. We found that in general, intermediate constraint exercises were as usable by students as less constrained formats, and engaged students in exploring complicated ideas and skills. Intermediate constraint formats contained more information on student thinking than multiple-choice, and in at least some cases showed a similar range of answers as more open-ended formats. We have evidence that intermediate constraint formats can have greater effects on student learning than less constrained formats, usually but not always in positive ways. Some intermediate constraint exercises do take longer to design than either high or low constraint exercises.

Based on this synthesis of evidence from student data, we will suggest where different levels of constraint are most appropriate for inquiry-based educational software.