PROBLEM SOLVING PROCESSES ASSESSED BY VERBAL RESPONSES

This study investigated the problem solving processes employed by fifth and sixth graders. The verbal responses of children engaged in problem solving were tape recorded and categorized. Several verbal response categories were identified from the literature and used to analyze children\u27s responses. They included surveying information, generating new information or a new hypothesis, developing information or testing one or more hypotheses, unsuccessful solution and self-reference or self-criticism. Thirty good and thirty poor problem solvers were identified on the basis of teacher ratings. Subjects were given practice tasks to facilitate thinking-out-loud behavior. The experimental tasks administered were two Electronic Mastermind problems. These tasks were presented on a small computer and subjects were required to discover a three digit code chosen by the investigator. The relevant dimensions to discover were the number itself and the position of the number in the code. The dependent variables were the proportions of the verbal response catagories and the frequency of paired verbal response category sequences. The results indicated that on one problem the good problem solving group generated and developed information to a significantly greater degree whereas the poor problem solving group used a higher proportion of unsuccessful solutions or errors. Comparable significant differences were not found between good and poor problem solvers on the second problem, but significant differences were found between good and poor problem solvers on both problems when comparing the observed and expected frequency of paired verbal response sequences. On Problem 1 the poor problem solving group frequently surveyed information for understanding and had frequent unsuccessful solutions. The good problem solving group surveyed information for understanding and then generated and developed one or two new hypotheses. The more efficient strategy was found to be generating two ideas at one time. On Problem 2 the problem solving strategy was different for both groups. The good problem solving group surveyed information more frequently initially to achieve understanding. Once understanding was achieved problem solving proceeded as it did on Problem 1. On the other hand, the poor problem solving subjects began to process information more productively on Problem 2 indicated by fewer errors. Additional analyses found that the females in the poor solving group used significantly fewer unsuccessful solution responses (errors) on Problem 2. Furthermore, the females in general used a higher proportion of responses involving generating new information on Problem 2 than on Problem 1. In this study different problem solving processes were employed by good and poor problem solvers on each problem administered. These results suggest that different problems may be solved by a unique combination of responses. Furthermore, the results indicate the importance of understanding in problem solving as well as the adaptive nature of the problem solver who is capable of changing strategies as the nature of the task changes. The implication is that more careful study of specific problems and processes is needed