The Effect of Logo (Turtle Graphics) on the Problem Solving Strategies Used by Fourth Grade Children

Abstract

The major purpose of this study was to determine the effect of learning the programming language Logo on the problem solving strategies used by fourth grade children while they solved mathematical word problems without a computer. A secondary purpose was to investigate whether children who had learned to program with Logo would regard mathematics more favorably as measured by a mathematics attitude survey. This eleven-week study was designed to teach Terrapin Logo (Turtle Graphics) to 15 out of a total of 30 fourth grade children. The study compared the audiotaped and written responses to a four-question problem solving posttest to determine which strategies were used by each group and whether the Logo group applied the strategies that they had been taught during their programming experience when solving non-computer problems. The Logo curriculum, written by the investigator, focused on teaching children the basic commands of the language while stressing the need to plan a procedure before beginning it, breaking a large idea into more manageable parts, guessing and then checking work and finally checking over the finished product. Some group instruction occurred, but individual or paired instruction was emphasized. Guided discovery ensured that strategies were being used and commands correctly applied. The statistical analysis consisted of computing means and standard deviations on individuals' attitude responses and a t-test was performed on these means. Chi-square critical values were computed for the process coding sequence, as well as t values. The following results were obtained from the study:

On the attitude measure significant differences were found only between the experimental and control groups on their general attidude toward school. Statistically significant differences occurred on several posttest problems using the strategies of checking and looking back, as well as making an error and then immediately correcting it. The investigator concluded that the strategies of checking and looking back over one's work and correcting an error were transferred by the experimental group when solving noncomputer word problems. Several important factors may have contributed to the lack of significant differences elsewhere. Among them are 1) short duration of the study, 2) confusion over application of computer strategies to mathematical problems, 3) limitations of process coding sequence, and 4) population size.