128
Portland
Wednesday afternoon
Session GD: PER: Student Reasoning
and Topical Understanding
Location: Pavilion East
Sponsor: AAPT
Date: Wednesday, July 17
Time: 2:40–4:20 p.m.
Presider: MacKenzie Stetzer
GD01:
2:40-2:50 p.m. Probing Inconsistencies in Student
Reasoning: Formal vs. Intuitive Thinking
Contributed – Nathaniel Grosz, North Dakota State University, Department of
Physics, Fargo, ND 58108-6050;
Mila Kryjevskaia, North Dakota State University
MacKenzie R. Stetzer, University of Maine
Even after targeted instruction, many students still struggle to productively
and consistently analyze unfamiliar situations. We have been designing
sequences of questions that allow for an in-depth examination of incon-
sistencies in student reasoning approaches. Our results indicate that even
those students who do possess the knowledge and skills necessary to ana-
lyze many challenging situations correctly often fail to utilize relevant ideas
and skills productively; students tend to “abandon” their correct formal
reasoning approaches in favor of more intuitive solutions (perhaps more
appealing to them at that moment). We will present results from sequences
of questions administered in introductory calculus-based physics courses.
GD02:
2:50-3 p.m. Physics Reasoning: Biases Toward the
Most Available Variable
Contributed – Andrew F. Heckler, Ohio State University, 181 W Woodruff
Ave., Columbus, OH 43210;
Abigail M. Bogdan, Ohio State University
It has been demonstrated in a number of everyday contexts that when
reasoning about causes of simple phenomena, people tend to only consider
explanations that are the most available in their memory. Thus, even if
people know that two factors can influence an outcome, they often only
consider the one that is most available. We demonstrate this phenomenon
in the physics education context. Specifically, when students are asked to
determine explanations for the variation of some quantity, such as the tip-
ping of a balance scale, or the mass density of a material, they tend to only
consider the most available variable that causes the variation, even in cases
when that variable is physically non-causal. However, when interviewed
further, students are often able to reason about alternative explanations and
other potential variables. We discuss a range of known student difficulties
in physics in terms of this reasoning bias phenomenon.
GD03:
3-3:10 p.m. Effects of Belief Bias on Causal Reasoning
from Data Tables
Contributed – Abigail M. Bogdan,* The Ohio State University, 191 West
Woodruff Ave., Columbus, OH 43210;
Andrew F. Heckler, The Ohio State University
Students often fail to draw valid conclusions from simple tables of experi-
mental data. Our research suggests that part of their struggle might be
caused by the influence of prior beliefs. In this study, students were asked
to either verify or construct a claim about a causal relationship between
several variables based on information presented in data tables. We found
that students demonstrated belief bias in the ways they chose to cite data,
frequently treating their own theories as a source of evidence to be supple-
mented by or illustrated with examples from data. Because of this tendency
to hunt piecemeal through the tables for supporting examples, contra-
dictory data was often simply overlooked. However, even when noticed,
data that contradicted their theories was often ignored, misinterpreted to
conform, or discounted in some way.
*Sponsored by Andrew F. Heckler
GD04:
3:10-3:20 p.m. Effects of Training Examples on
Understanding of Force and Motion
Contributed – Daniel R. White, The Ohio State University, 191 W Woodruff
Ave., Columbus, OH 43210;
Andrew F. Heckler, The Ohio State University
We examined the effects of various kinds of training tasks on student
responses to questions about the relationship between the directions of net
force and velocity, and between acceleration and velocity in one dimension.
The four training conditions were constructed in a 2x2 design (abstract
vs. concrete contexts) x (acceleration-velocity vs. force-velocity question
types), and a control (no training) was also included. We found that all
training conditions significantly improved performance on all question
types compared to control, however acceleration-velocity training resulted
in higher performance on all question types compared to force-velocity
training. Additionally we found that the degree of abstraction of the
training (that is, the number of concrete details included in the example)
has no significant effects on student scores. These results are consistent
with hierarchies of student understanding of force and motion in previous
works, which we also discuss.
GD05:
3:20-3:30 p.m. Effectiveness of Computer-based
Training on Vector Products
Contributed – Brendon D. Mikula, The Ohio State University, 191 W. Woodruff
Ave., Columbus, OH 43210;
Andrew F. Heckler, The Ohio State University
Computer-based training on dot products and cross products was given to
N=223 students in an introductory level, calculus-based electromagnetism
course at The Ohio State University. The level of feedback in training was
varied as follows: no training (control), correctness feedback, correct-
ness and correct answer feedback, correctness and explanation feedback.
Training lasted for approximately 10 minutes, and a paper assessment was
given immediately afterwards. This assessment consisted of both arrow
format questions, similar to those in the training, and conceptual/transfer
questions related to vector products selected from the Vector Concept Test
(Zavala & Barniol). All training conditions significantly outperformed
control on both question types (d > 0.46). Consistent with VanLehn’s
interaction plateau hypothesis, high-level feedback was significantly more
effective than low-level feedback for arrow notation questions (d = 0.35)
and no significant difference was observed between the high-level feedback
conditions for either question type.
GD06:
3:30-3:40 p.m. Student Reasoning on Gravitational and
Electrostatic Potential Energy
Contributed – Beth A. Lindsey, Penn State, Greater Allegheny, 4000 Univer-
sity Drive, McKeesport, PA; 15132
Potential energy is a conceptually rich topic presenting many difficulties
for students. Recent research has identified many difficulties relating to
work, energy, and systems. Failure to reason correctly about potential en-
ergy may underlie many of these difficulties. I will describe an investigation
into student understanding of potential energy as typically presented in the
context of universal gravitation or electrostatics. I will discuss the connec-
tions between student understanding of potential energy in mechanics and
their subsequent performance in electricity and magnetism. I will present
data from written questions and from one-on-one student interviews,
and discuss the implications these data have for instruction on energy in
introductory courses.
GD07:
3:40-3:50 p.m. Students’ Initial Representations of
Light in College Physics
Contributed – Craig C. Wiegert, University of Georgia, Department of Physics
and Astronomy, Athens, GA 30602-2451;
Cameron Zahedi, University of Georgia
We report on college physics students’ prior diagrammatic knowledge
about light propagation and optics. At the beginning of the second semes-
