Identifying students' problem solving strategies in a programming task
Abstract
This research study seeks to nd problem solving and programming strategies used by students when approaching a solution to a programming task. It also studies how the programming strategies used by students relate to their performance on the test. Information processing theory coupled with problem solving search strategies was chosen as a theoretical framework. A qualitative study was conducted on twenty students answering a laboratory exam and their strategies were noted, codifed and analyzed to find themes and patterns. It was found that the students who succeeded were very good with problem decomposition and goal setting. It was also noted that students, irrespective of their skill level identifed all the intermediate stages of a problem solution before formulating the complete solution. However the students didn't have such a plan and dealt with sub problems as they face it, even before identifying all the intermediate stages when dealing with unknown problems. The students in low and average performing category were found to have problems with variables and datatypes. The low performing students were found to have a lacking model of the program and computer and were not successful debuggers. Students with the best debugging skills were found to be successful. This study underscores the importance of a viable mental model of a computer and program. It also encourages the instructors to instruct ecient ways to subgoal a given problem and solve it. Debugging strategies and tools are also identifed as the main characteristic which separates high performers from other categories thereby advocating more emphasis in this area by the instructors
Degree
M.S.
Advisors
Magana, Purdue University.
Subject Area
Educational psychology|Computer science
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.