Using analogies to improve elementary school students' inferential reasoning about scientific concepts

School Science and Mathematics, Mar 2001 by Yanowitz, Karen L

Performance on inference questions. Responses were scored as correct if they described a consequence that might occur as a result ofthe changed situation. For instance, students who were asked, "What would happen if the mitochondria stopped working?" were given credit for responses such as, "You would be very weak," or "You would probably die because your heart wouldn't get any energy. " Scores from each question were summed to obtain a total reference score. Therefore, scores could range from 0 (no correct answers) to 4 (correct answers for all four questions). The author and an undergraduate assistant independently scored all students' responses. Percent agreement on answers to each inference question ranged from 93% to 100% for each condition at each grade. Any disagreements were easily solved through discussion.

Table 1 also presents scores at each grade as a finction of text type. A 2 (text type; analogy, non analogy) x 2 (grade; fourth, sixth) ANOVA revealed that students receiving analogical texts answered significantly more inferential questions correctly (M = 2.8, SD = 1.1) than students receiving nonanalogical texts (M= 2.0, SD = 1.0), F(1,49) = 7.7,p

The results from Experiment 1 revealed that elementary school students benefitted from receiving analogies when learning science. Furthermore, the analogies had a very specific effect on students' performances. Text type did not affect students' recall of the basic factual knowledge about the science concepts. In contrast, students who received analogies showed higher levels of inferential reasoning about the science concepts than students who received nonanalogical texts.

Experiment 2

Experiment 1 revealed that analogies helped elementary school students reason inferentially about various science topics. This result is comparable to studies conducted with adult learners (Donnelly & McDaniel, 1993; Halpern et al., 1990; Iding, 1993; McDaniel & Donnelly, 1996). Learning the concept with an analogy may allow the student to develop a better understanding of the basic principles of the concept. Therefore, these students were better able to predict the results of changes in the domain than students who learned the concept without an analogy.

An instructional analogy explicitly compares the common structural relations between a source and target domain. Thus, the main idea or general principle governing the domains is presented twice. For instance, considerthe mitochondria and power company analogy. Both domains describe how one thing (mitochondria or power company) sends out energy to make other things (parts of the body or appliances in the house) function. Therefore, the structural relations are described once in the context of the science concept and once in the context ofthe source domain. In contrast, the structural relations are presented only a single time when the science concept is described in the nonanalogical text. The repetition of the structure may be a factor in the higher inferential reasoning shown by students who received analogical texts. Even though no difference was found in the recall ofthe main idea, perhaps reading the structure twice enabled a better understanding of the general principle, thereby facilitating inferential reasoning about the effect of changes in the domain. The primary goal of Experiment 2 was to explore this factor as a possible influence on the improved inferential reasoning for students who read analogical texts.