Assessing spelling skills and strategies: A critique of available resources

Abstract

In this paper, we discuss a variety of spelling tests that are used to assess developmental spelling difficulties. We differentiate between tests that are valuable tools to monitor spelling development and spelling tests that should be used to further assess children who are not making sufficient progress in the mainstream classroom. We recommend the use of several tests in order to achieve these assessment goals. This paper is a practical guide for clinicians and teachers concerned with the assessment (and treatment) of poor spellers.

Acknowledgements

During the preparation of this paper, Lyndsey Nickels was funded by an NHMRC Senior Research Fellowship. This research was in part supported by an Australian Research Council grant to Anne Castles.

Notes

1. Note that letters between // represent sounds. Letters between ‘’ represent letters.

2. Note that often, the alphabetic/phonetic stages in the developmental models are equated to the acquisition of the sound-letter procedure while the orthographic stages are acquisition of the whole word route.

3. While Dual Route Theory acknowledges the interaction between the two procedures it can also explain patterns of performance where orthographic knowledge is better developed than would be predicted based on phonological knowledge as is the case in phonological dyslexia/dysgraphia (i.e. poor nonword reading/spelling, good word reading/spelling). Stage theory has difficulties explaining ‘phonological dyslexia/dysgraphia’.

4. Note that these tests currently do not have Australian norms. Hence, care must be taken when interpreting results.

5. While there is evidence to suggest that phonemic segmentation not segmentation into onsets and rimes best predicts later spelling abilities (Nation & Hulme, 1997), we are not aware of research that ascertains whether the sound-letter route can work without the ability to explicitly segment sound strings into smaller units. Hence, in the classroom and for training, explicit ability to identify single sounds and relating them to letters is needed.

6. Note that only the QUIL has norms from Australian children. Norms from overseas student populations, of course, have to be interpreted with caution.

7. Test instruments are checked for their reliability, i.e. how much does a given student's test score vary at different testing times. Test-retest reliability is expressed as a number between 0 and 1. The higher the number the better the reliability.

8. The only processing component that is not discussed in detail in the paper is the Graphemes/Graphemic buffer. Difficulties at this level are reflected in, for example, length effects. However, isolated problems at this level have not been reported in children to this date.

9. Note that people with very pure deficits that only affect the sound-letter component will not be identified by either the BSTS or the SWST. However, it seems unlikely that such a pure impairment would result in marked functional deficits (see, e.g., Howard & Best, 1996).

10. We recommend the QUIL rather than ITPA-3 because the ITPA-3 has floor effects and uses a somewhat unusual task. The QUIL's advantage over the Woodcock-Johnson ‘Sound Spelling Test’ is that the QUIL has Australian norms. For teachers and clinicians in North America, the recommendation might be different.

11. Also, a lot of children in Australian schools seem to be very familiar with the items as the test is used so widely.

12. Some of the items can be spelled correctly by applying the most frequent sound-letter rules, though, e.g. name.

13. Regular was defined as the most frequent way to spell a certain phoneme according to the Perry list (Perry et al., 2002).

14. Note that a lot of the tasks we described in this article exist for adults, e.g. in the Psycholinguistic Assessment of Language Processing in Aphasia (Kay, Lesser & Coltheart, 1992) or The Johns Hopkins University Dysgraphia Battery (Goodman & Caramazza, 1985). However, these tests were designed for adults, hence there are no paediatric norms.