Abstract
A comparison of the effectiveness of 5 presentation methods for the teaching of spelling was made with 359 children, where the words to be learned were presented in either isolation or in context, and where the method of learning involved either the study of the whole word, the 'hard' spot or the individual letters of each spelling. A battery of tests was given to the subjects in order to obtain measures of some of the individual characteristics thought to be relevant to the investigation. The administration of the test battery and working through the spelling learning programme covered a period of four months. At the beginning of the investigation the age range of the children was from 9 years 2 months to 10 years 3 months. Results showed that a systematic approach to the learning of words leads to improvement in spelling performance. Analyses of variance revealed;-1. Significant differences between the average gain scores of the non-manual and manual social class groups, the latter making the larger gains. 2. Significant differences between the five presentation methods; the five treatments were as follows: - Presentation: Method Presentation Method of Learning: P List form of isolated words. Whole word; Q List form of words + meanings. Whole word; R Initially in a story. Then words + meanings + use in sentences. Whole word. S. Initially in a story. Then attention to 'hard spots' + meanings + use in sentences. Part word T. Initially in a story. Then attention to individual letters + meanings + use in sentences. Part word. The results indicated that an approach which presents the words in context and utilises a whole word method of learning (Presentation method R), is the most effective one to use with a whole class of mixed social class groups. 3. There were no significant differences between methods of learning which (a) used an oral and (b) a written response mode. 4. Highly significant differences were found between a learning task of four words and that of six words daily, with the latter giving consistently higher gain scores for presentation methods and social class groups. Using standardised test results as measures of factors within the subjects, and the average pre-test, post-test and gain scores as the dependent variables; three series of regression analyses were carried out. Results showed that by far the most powerful predictor of spelling performance was the spelling age. The highest additional contribution to the pre-test and gain scores being made by the verbal reasoning quotient, and to the post test scores by the silent reading age. The child variables of spelling age, verbal reasoning quotient, reading age, and the recall of words, (using visual and auditory presentation), measured immediately before the beginning of the learning of the sets of words, showed high inter-correlations and also correlated highly with the three dependent variables.