A recent study, “Measures of Kindergarten Spelling and Their Relations to Later Spelling Performances” by Rebecca Treiman, Brett Kessler, Tatiana Cury Pollo, Brian Byrne, and Richard K. Olson, and published in Scientific Studies of Reading Vol.20 (349-362), administered spelling tests to kindergarteners and two years later as second graders to measure spelling proficiency levels. These spelling tests were scored using 8 different scoring methods including orthographic (letter based) and phonological plausibility. Phonological plausibility credits incorrect spellings if the spelling make phonological sense; for example, /kat/ for the word cat would not be marked incorrect because of the phonological relationship between /k/ and /c/. The variety of scoring methods was used to determine which method was most accurate in showing kindergarten to second grade progress-- whether positive progress or not. Would a student who spells words orthographically incorrect but phonologically plausible in kindergarten develop into good orthographic speller in second grade because they showed phonological understanding in kindergarten?
Participants were made up of 158 American and 216 Australian students who took a 10-word spelling test at the end of kindergarten and another, longer, test at the end of Second grade. These spelling tests were then scored with two binary (fully correct or incorrect) methods-- orthographic correctness (standard spelling) and phonological plausibility-- and six nonbinary methods that granted partial credit based on letter sequences, phoneme distance, and a method based on a mixture of crediting both letters and phonemes.
The results of this study showed that while all measurement methods used in this study “correlated significantly with second-grade spelling performance,” they differed in degrees of correctness. Phonological plausibility scores were the lowest of accuracy with second-grade spelling performance. The mixed methods were about 57% correct in indicating later spelling performance, and orthographic correctness was most correct at 62-63%. Results of this study also show that students with a lot of letter reversals in kindergarten spelling tests correlated with poor spellers in second grade-- supporting the use of letter reversals for an early indication of dyslexia and other language-based learning disabilities.
These results indicate that, surprisingly to the authors, measurement of orthographic correctness is more telling of later spelling performance than phonological plausibility measurements, which suggest that even at the kindergarten age students have a good understanding of “conventional [spellings] for specific words or specific positions in words.” Children who have a good orthographic understanding, then, will be more advanced in later spelling performances. This perhaps means that phonological instruction should come earlier than this age, before students learn to simply memorize conventional spellings.