The effects of computer programming on children's higher-level cognitive processes: a review of the research on logo
Francine Ruth Doughty Palmeter · 1991
Seymour Papert, the creator of the Logo programming language, claimed that interaction with Logo and the classroom culture surrounding its implementation would effect four different areas of cognition. He claimed the child would (1) indirectly acquire mathematical knowledge and intuition; (2) acquire combinatorial skills at an age prior to the formal operational stage of development; (3) engage in greater levels of self-referential thought; and (4) acquire general problem-solving skills. Researchers quickly sought to verify these claims. Because of varied results, subsequent reviewers of this research concluded that Logo was not an effective tool for the classroom. There were serious weaknesses with these reviews and therefore a systematic review of the Logo literature was undertaken. Using the procedures of meta-analysis, the effect of Logo on the four areas of cognition was examined. A fifth area, homemade testing, was added when it was thought that effect sizes (ESs) might be artificially increased because the experimenter was not only instructing Logo, but also was creating the test. This meta-analysis departs from other meta-analyses in the following ways: (1) Experts were used to classify all dependent variables into one of the five categories; (2) A new formula was used to weight ESs by the number of comparisons from each study and its methodological rigor; (3) Decision rules were carefully followed so that homogeneity of ESs might be determined; and (4) Methods were outlined clearly so that they could be replicated or modified. The results of this literature review contradict some findings of previous literature reviews that question the value of Logo instruction. Logo had a positive impact on problem-solving skills and metacognitive ability. These effects were most pronounced when (a) the instructor had mastery of the language; (b) a guided discovery method of instruction was employed; and (c) the child decided the project to be programmed. These data suggest that it would be premature to devalue Logo as an instructional tool.