Text editing in chemistry instruction

Abstract

In two experiments, differential performance onchemistry problems was obtained for twotraining strategies: text editing andconventional problem solving. Text editingrequires students to scan the text of problemstatements and specify whether it providessufficient, missing or irrelevant informationfor solution. It was hypothesized that textediting, which emphasizes gaining familiaritywith schematic knowledge, would lead to higherachievement than conventional problem solving.Experiment one indicated that text editing wassuperior to conventional problem solving inlearning to solve molarity and dilutionproblems. In particular, students who weretrained in text editing skipped someintermediate steps while solving molarityproblems. In contrast, using stoichiometryproblems, experiment two showed that students whowere trained in text editing performed worsethan students given conventional problems tosolve. An error analysis suggested that becauseof its failure to direct students' attention tothe coherent problem structure in the firstinstance, text editing has no advantage overconventional problem solving in the domain ofstoichiometry problems. It was concluded thatthe suitability of a text editing trainingstrategy depends on the learning materials.