|
Article Excerpt
Abstract. A spelling training was implemented in three subgroups of Dutch children (N = 33) with spelling disabilities: (a) children with spelling problems of normal or above-normal intelligence; (b) children with spelling problems of normal or above-normal intelligence who also exhibited severe externalizing behavioral problems; and (c) children with spelling problems of relatively low intelligence. Two types of words were trained (N= 24): words with ambiguous sound-spelling relations and words with complex consonant clusters. The training was effective in all subgroups. The spelling performance on both types of words improved from pretest to posttest, and was on the whole sustained on the retention test. The only exceptions were the children with spelling problems and severe behavioral problems, whose performance on words with ambiguous sound-spelling relations dropped from posttest to the retention test, even though it remained above pretest levels. Theoretical and practical implications for the spelling instruction of subgroups of children with spelling problems on subclasses of words are discussed.
**********
Mastering the intricacies of written language is a major challenge for beginning readers and spellers. In alphabetic languages, beginning readers and spellers must learn the relationships between phonemes (sounds) and graphemes (spellings). In some alphabetic languages, for example, Finnish and Spanish, these sound-spelling relations are rather predictable. That is, when presented with a phoneme a speller can almost always predict which grapheme is connected to the phoneme. Other alphabetic languages, like English, French, and Dutch, are highly inconsistent in terms of the sound-to-spelling relationship. As a result, when presented with a phoneme, a speller from these languages most likely has to choose from two or more possible graphemes. With regard to the connections from sound to spelling, Finnish and Spanish are shallow orthographies whereas English, French, and Dutch are deep orthographies. This distinction between deep and shallow is based upon the connections from sound to spelling rather than spelling to sound. With respect to its spelling-to-sound relationship, Dutch and French (Ziegler, Jacobs, & Stone, 1996) are relatively shallow, whereas English is deep (Stone, Vanhoy, & Van Orden, 1997). Generally, in most alphabetic orthographies spelling is more difficult than reading because of the lower consistency from sound to spelling than from spelling to sound (Bosman & Van Orden, 1999).
Predictable sound-spelling relations refer to a one-to-one correspondence between phonemes and graphemes. For example, the Dutch phoneme [m] is almost always spelled M. More often, however, at least in Dutch and English, sound-spelling relations are ambiguous and unpredictable. For example, the Dutch phoneme [Ei] can be spelled with two graphemes "ij" and "ei," and the Dutch phoneme [Au] can be spelled with four graphemes "ou," "ouw," "au," or "auw." Given the difficulty of learning to read and spell, it is not surprising that some children have problems acquiring reasonable reading and spelling skills.
Teaching spelling skills to children with spelling disabilities is important for several reasons (cf. Fulk & Stormont-Spurgin, 1995; Graham, 1999). First, at least in literate societies, spelling skills are often conceived as an indicator of a person's linguistic or even intellectual abilities. As a result, children who make a relatively high number of spelling errors in their essays receive lower grades than children who make few errors, even when the overall quality of the essays is alike (Marshall & Powers, 1969, in Graham, 1999). Second, difficulties in finding the correct spelling for individual words may interfere with the pleasure to write (Berninger, Vaughan et al., 1998), thereby potentially constraining the development of higher-order writing skills involved in text writing, such as conceptualization and planning (Graham, 1999; MacArthur & Graham, 1987).
If untreated, spelling problems may persist throughout the elementary grades (e.g., Juel, 1988). Empirical studies indicate that children with spelling disabilities tend not to acquire spelling skills naturally through reading and writing (see Graham, 1999, 2000, for reviews; cf. Krashen, 1989). Rather than relying on a natural learning approach, children with spelling disabilities appear to need formal spelling instructions to attain spelling skills. This conclusion is substantiated by teachers' experiences that explicit spelling instruction is imperative to overcome spelling problems (e.g., Berninger, Vaughan, et al., 1998; for a review, see Fulk & Stormont-Spurgin, 1995).
In conclusion, developing and implementing effective spelling instruction for children with spelling disabilities is important. Although their spelling problems appear persistent, teaching spelling skills to these children is feasible, and instruction should preferably start at a relatively young age.
In the present article, we report on a spelling training program implemented with three subgroups of children with spelling disabilities attending different schools for special education. Additionally, the study examined the merits of the training for learning to spell different types of words; namely, phoneme-to-grapheme consistent and phoneme-to-grapheme inconsistent words. We will first outline the spelling training and its underlying assumptions regarding spelling cognition.
VISUAL-DICTATION TRAINING
In our spelling training, which we called visual dictation, children are visually presented with a word that they are told to study carefully. Next, the word is covered and they are asked to write the word in their notebook. The word is then shown again and the children check its spelling. In case of a misspelling, they correct their own writing and rewrite the word (while the word is visible).
In previous studies, we implemented a precursor of the visual-dictation training in a group of typically achieving spellers in first grade (Bosman & van Hell, 1999; van Leerdam, Bosman, & Van Orden, 1998). It comprised all the instructional elements described above but without a self-correction phase. The performance of the typically achieving students in the visual-dictation condition was compared with that of four groups of typically achieving spelling peers who followed one of four training procedures: (a) read each word aloud (reading); (b) copied each word into a notebook (copying); (c) read each word aloud, after which the teacher covered the word, presented the target and a nontarget grapheme and asked the child to circle the correct grapheme (grapheme selection); and (d) read each word aloud and spelled each letter of the word from memory (oral spelling). All children were trained on the same series of words, each containing one targeted ambiguous sound-spelling relation. After training, children in the visual-dictation condition made fewer spelling errors on a dictation test than children in any of the other training conditions; this advantage could not be explained by differences in time on task.
The visual-dictation training integrates four instructional principles that have been identified as efficacious: Writing from memory, kinematics of writing/ practicing targeted spelling difficulty in whole word, and immediate feedback by means of self-correction. We will discuss the merits of each of these principles in more detail.
Writing From Memory
When writing a text, writers typically produce the spelling of words from memory. Therefore, training in which writing words from memory is practiced may yield higher spelling gains than training in which the to-be-written words remain visible. This is corroborated by the findings of several studies showing that writing words from memory during training leads to better results than simply copying words from a sheet of paper. In a spelling training with beginning readers, Roberts and Ehri (1983) observed that children who were trained on words they had to imagine achieved higher scores than children who were trained on words they could see on a piece of paper. This finding was obtained for both skilled and less skilled readers. Bosman and de Groot (1992) also tested beginning readers and found that spelling words from memory was more effective than copying words, particularly for children with spelling problems. These effects were replicated in subsequent spelling-training studies (Bosman & van Hell, 1999; van Leerdam et al., 1998).
Kinematics of Writing
The motor activity involved in handwriting is commonly assumed to benefit spelling performance: Handwriting enables children to benefit from motor feedback in learning to spell (e.g., Graham, 1999; Hulme & Bradley, 1983) and may contribute to strengthening orthographic knowledge (Berninger, Abbott et al., 1998). The importance of handwriting is supported by a study by Cunningham and Stanovich (1990), who examined the spelling performance of first graders. All children were presented with cards containing printed words that were read by the experimenter and subsequently repeated by the child. The child then had to copy the word by using either handwriting, letter tiles, or a computer keyboard. A subsequent dictation test showed that children in the handwriting condition performed better than children who had used letter tiles or the computer keyboard.
Whether children with spelling disabilities also benefit from the kinematics in handwriting remains unsettled. In spelling instruction with children with spelling disabilities, Berninger, Abbott et al. (1998) observed that children who had written by hand outperformed their peers who had used the computer, albeit only on relatively easy words. However, Vaughn, Schumm, and Gordon (1993) observed no benefits of handwriting over computer writing in their spelling training for children with learning disabilities (similar results were obtained with children without spelling disabilities). Still, children thought they learned spelling best when writing by hand, as revealed by a post-training interview.
The merits of kinematics have also been recognized in Orton-Gillingham's multisensory techniques, in which the student first traces (or sometimes writes) each letter in a word while simultaneously saying its name, and subsequently learns to read and spell (see Hulme & Bradley, 1983). The difference between our visual-dictation task and the standard multisensory technique is that in the latter each letter is connected to its sound through kinematics.
Practicing Targeted Spelling Difficulty in Entire Word
Many spelling methods, at least in the Netherlands, focus on a specific, typically ambiguous, sound-spelling relationship in a word. In learning the spelling of the word "zout," for example, Dutch children have to explicate the ambiguous grapheme [Au]. However, focusing on the ambiguous grapheme largely ignores the context of the whole word in which graphemes are embedded. This is...
|
