The effect of instruction method and relearning on Dutch spelling performance of third- through fifth-graders

In this study, we compared two instruction methods on spelling performance: a rewriting instruction in which children repeatedly rewrote words and an ambiguous property instruction in which children deliberately practiced on a difficult word aspect. Moreover, we examined whether the testing effect applies to spelling performance. One hundred eighty-six Dutch elementary-school students (grades 3, 4, and 5) participated in this study. A mixed design was used in the present study, with age group and instruction as between-subject variables and relearning as a within-subject variable. We showed that after a 2-day retention interval, the rewriting condition outperformed the ambiguous property condition on spelling performance in all grades. The effect of relearning type was not significant nor was the instruction x relearning interaction. An error analysis showed that relative to the rewrite instruction, the ambiguous property instruction led to more errors on the non-practiced part of the words. By contrast, the rewrite instruction and ambiguous property instruction did not differ with respect to the errors on the practiced part of the words. The findings provide strong evidence for the superiority of a rewriting study instruction over an ambiguous property study instruction with respect to the performance on a delayed spelling test. Results from the conditional error analyses suggest that the beneficial influence of rewriting emerges because rewriting requires children to process the whole word rather than only a part of the word.