The Effect of Context on Student Performance on a Homework-Style Problem

Andrew R. Sloboda

Download Paper | Permalink

Conference: Middle Atlantic ASEE Section Spring 2021 Conference
Location: Virtual
Publication Date: April 9, 2021
Start Date: April 9, 2021
End Date: April 10, 2021
Page Count: 10
DOI: 10.18260/1-2--36325
Permanent URL: https://peer.asee.org/36325
Download Count: 338

Request a correction

Paper Authors

biography

Andrew R. Sloboda Bucknell University orcid.org/0000-0002-0729-3278

visit author page

Andrew Sloboda is an Assistant Professor at Bucknell University where he teaches a variety of mechanics-based courses, including statics, solid mechanics, fluid mechanics, dynamics, system dynamics, and vibration. His research interests lie primarily in the fields of nonlinear dynamics and vibration.

visit author page

Download Paper | Permalink

Abstract

Homework, in the form of quantitative problems, is a ubiquitous part of the engineering learning experience. Solving problems on assignments helps students to develop problem-solving skills, practice applying learned material to relevant situations, and can reveal issues with conceptual understanding.

Problems can be designed with different levels of context, that is, different amounts of backstory, visual representations, and additional information (whether pertinent or not). It stands to reason that the level of context in a problem might be beneficial or detrimental depending on a learner’s level of understanding and the instructor’s learning goals. One might hypothesize, for instance, that a problem with only basic and pertinent information might be beneficial for someone wrestling with a particular analysis for the first time, whereas a problem that better resembles a real-life scenario might provide a learner the opportunity to practice far transfer with a skill that is nearly mastered.

In this work-in-progress, student performance on a problem-solving task is studied. Two groups of students are presented with fundamentally the same problem. However, one group’s version of the problem contains only the information necessary to solve the problem, while the other group’s version contains additional details in the form of further narrative description and a figure. Each student is asked to rate how well they can connect the problem to principles and methods previously discussed in class and how confident they are that they can solve the problem. They then time themselves while solving the problem.

Analysis of student responses shows that students presented with even a minimal level of additional problem context will report decreased confidence in their ability to connect the problem to past learning and in their ability to solve the problem. Additional context also increases the time students take when solving the problem, but does not seems to impact performance as measured by arriving at a correct solution.

This suggests the design of problems should be tailored depending on learning goals. Additional context should only be included when further transfer of learning is expressly desired.

Citation
Format

Sloboda, A. R. (2021, April), The Effect of Context on Student Performance on a Homework-Style Problem Paper presented at Middle Atlantic ASEE Section Spring 2021 Conference, Virtual . 10.18260/1-2--36325

TY - CPAPER
AB - Homework, in the form of quantitative problems, is a ubiquitous part of the engineering learning experience. Solving problems on assignments helps students to develop problem-solving skills, practice applying learned material to relevant situations, and can reveal issues with conceptual understanding.

Problems can be designed with different levels of context, that is, different amounts of backstory, visual representations, and additional information (whether pertinent or not). It stands to reason that the level of context in a problem might be beneficial or detrimental depending on a learner’s level of understanding and the instructor’s learning goals. One might hypothesize, for instance, that a problem with only basic and pertinent information might be beneficial for someone wrestling with a particular analysis for the first time, whereas a problem that better resembles a real-life scenario might provide a learner the opportunity to practice far transfer with a skill that is nearly mastered.

In this work-in-progress, student performance on a problem-solving task is studied. Two groups of students are presented with fundamentally the same problem. However, one group’s version of the problem contains only the information necessary to solve the problem, while the other group’s version contains additional details in the form of further narrative description and a figure. Each student is asked to rate how well they can connect the problem to principles and methods previously discussed in class and how confident they are that they can solve the problem. They then time themselves while solving the problem.

Analysis of student responses shows that students presented with even a minimal level of additional problem context will report decreased confidence in their ability to connect the problem to past learning and in their ability to solve the problem. Additional context also increases the time students take when solving the problem, but does not seems to impact performance as measured by arriving at a correct solution.

This suggests the design of problems should be tailored depending on learning goals. Additional context should only be included when further transfer of learning is expressly desired.
AU - Andrew R. Sloboda
CY - Virtual
DA - 2021/04/09
PB - ASEE Conferences
TI - The Effect of Context on Student Performance on a Homework-Style Problem
UR - https://peer.asee.org/36325
DO - 10.18260/1-2--36325
ER -