Lexikala analyser av muntlig, tangentbordsskriven och dikterad text producerad av barn med stavningssvårigheter

Abstract

For barn med stavningssvarigheter ar skriftsprakande en stor utmaning och behovet att hitta nya redskap for att understodja deras skriftspraksutveckling ar darfor stort. Att skriva med taligenkanning, diktering, skulle kunna underlatta skrivprocessen eftersom dikteringsverktyget reducerar kraven pa skribenten att stava sjalv, och detta skulle darmed kunna frigora kognitiva resurser till annan textbearbetning. Att diktera text innebar emellertid inte med nodvandighet en enklare textproduktion; dels behover texten folja skriftspraksnormer trots att den produceras muntligt, dels behover skribenten hitta strategier for att hantera tillfallen nar verktyget inte uppfattar talet korrekt. I foreliggande studie undersoks text producerad av barn med stavningssvarigheter. Lexikala egenskaper, textlangd och stavfel studeras i muntlig, skriven och dikterad text. Analyserna visar att de dikterade texterna innehaller farre stavfel an de skrivna samt att bade skriven och dikterad text har hogre lexikal densitet och hogre andel langa ord jamfort med muntlig text. Detta antyder att deltagarna klarar att diktera texter med skriftsprakliga egenskaper. Daremot skiljde sig dikterad text varken fran muntlig eller skriven text gallande textlangd eller lexikal diversitet. Var slutsats ar att dikteringsverktyg kan anvandas for att underlatta skrivprocessen for barn med stavningssvarigheter, men det behovs vidare forskning som studerar hur olika undergrupper producerar text med dikteringsverktyg. Abstract Lexical analyses of oral, typed and dictated text produced by children with spelling difficulties For children with reading and writing difficulties, writing, in general, and spelling, in particular, is a significant challenge. Writing with speech recognition (dictation) can facilitate the writing process since the dictation tool removes the need to spell, freeing resources for other processes. Dictating, however, may not always provide a solution, since written text should follow the conventions of written language, even when produced orally. In addition, writers need strategies if the tool does not recognize the spoken words.In the present study, children with spelling difficulties produced keyboard-written and dictated texts, and an oral account. The analyses showed that dictated texts, as expected, contained fewer spelling errors than keyboard-written texts, and that both of these had a higher lexical density than the oral accounts. This suggests that students are able to dictate using written language conventions. However, the dictated texts did not differ from the oral and keyboard-written texts in terms of length of text and proportion of long words. Our conclusion is that the dictation tool can be used to facilitate the children’s writing process, but further research is needed to see if training in the use of the tool would bring more substantial benefits.