Student Responses and Attitudes Concerning the Use of Physical Models in a Large Classroom Environment

As part of nine, large (n>150), general education, undergraduate earth science classes, students used hands-on, physical models to explore abstract concepts such as the seasons, plate tectonics, rock characteristics, half-life, relative time, fronts, and past climates. Students worked in groups of three or four, and trained student observers recorded student interactions during modeling activities. Sixty students from six classes were interviewed to assess their reactions to these models. In addition, conceptual retention was assessed by asking conceptest questions to 26 students from three classes. Classroom observations indicated that higher achieving students referred to the models as platforms for discussion and peer instruction. If students struggled to understand the directions, they focused on rote completion of the task, but if instructions were clear and students were told how the model aided their learning, they focused on using the models to understand the relevant concepts. Groups that worked well together engaged in deeper conceptual discussions than groups who refused to cooperate. Analyses of interview results indicated that the students overwhelmingly preferred using the models to lecture only – even the active lectures utilized in these classes. Students who considered the models to be helpful reported positive dynamics within their groups while students voicing a negative opinion of the models reported being part of a non-functioning group. Reasons for preferring the models included a hands-on learning style, increased attention to class content, and construction of visual memory aids that could be used during the exam. Quantitative data comparing model-based to non-model based instruction were inconclusive. Comparisons between pre-instruction and post-instruction conceptest questions varied between semesters, suggesting that factors such as model implementation or group dynamics may have influenced the results. Overall, these models worked best as a common reference point that exposed common misconceptions and provided an opportunity for conceptual change.