Using forward inferencing as an indicator of problem solving skill in U.S. and Indian engineering undergraduates

With rising concerns about U.S. competitiveness in a global workplace, questions typically center on adequacy of training. What are the parameters of achievement, what benchmarks can be established as indicating competent and proficient student performance? Research on problem solving has shown that the specific order in which a person generates equations in a solution is indicative of his or her level of expertise. Experts apply the process of forward inferencing. Novices apply the process of backward inferencing. Forward inferencing requires a deep understanding of the problem. This understanding is activated immediately, either through recalling that type of problem from past experience, or through reasoning about the problem before generating equations. Students' problem solving protocols were analyzed to determine if they behaved like experts or novices. The data consisted of paper-and-pencil solutions and videorecordings of engineering freshmen and sophomores who were asked to think aloud as they solved typical statics problems. Data from U.S. students suggested that freshman-sophomore undergraduate students did not use forward inferencing. In contrast to the U.S. data, students at an Indian Institute of Technology clearly used forward inferencing and showed that beginning undergraduate students can achieve the deep problem solving insight characteristic of experts. The U.S. and Indian data include quantitative and qualitative evidence. The distributions of forward versus backward inferencing are reported. Curriculum and cross-cultural differences are considered, in part, in accounting for the differences between U.S. and Indian students. The value of this research to U.S. engineering curricula is that it provides clear pedagogical benchmarks for classroom instruction. Undergraduate engineering programs are intended to develop effective problem solvers. A problem-centered approach that requires students to solve practice problems from a textbook is common in many courses. Data collected from two U.S. universities0 found that solving textbook problems was the most frequent and time-consuming learning activity outside of attending course lectures. These demands of time and effort are placed on students for good reasons. The goal of problem-solving practice is to develop basic knowledge and procedural skill, and, importantly, to ultimately develop students with a firm grasp of conceptual principles and the ability to think analytically and critically about problems in a domain. Within current instructional practices, it is not always clear how to define proficiency or expertise in undergraduate problem solving. Research has shown that engineering curricula orient students to final solutions, not necessarily to the process followed in order to achieve the solution.1 Relatedly, feedback from instructors is often oriented toward the accuracy of the final solution and not the strengths and weaknesses in students' conceptualizations of problems.1 The present work attempts to develop a measure of definiteness of what it means to be a skilled problem solver as an undergraduate student. It also confronts the absence of clear benchmarks or expectations regarding students' facility and proficiency in problem solving. Finally, this paper presents an experimental paradigm and methodology for systematically exploring the parameters of skill development in engineering undergraduates.