In this paper we present a study of some spelling error types that Dutch primary school children made in the dictations and in the free or themed texts they contributed to the BasiScript corpus, i.e. a corpus comprising child written output produced between 2012 and 2015. The present article first briefly describes the corpus. Then it presents an analysis of the spelling errors that occurred in a selected set of words in the dictations regarding diphthongs (in grades 2 and 3) and verb forms (in grades 4 and 5) – which are notoriously difficult to spell for these age groups. In our analysis we investigate whether the frequencies of the words in the BasiLex corpus (a corpus of child written input) predict the spelling errors and whether there is a correlation between number of incorrect spellings of the words in the dictations and in the free texts and themed texts of the respective grades.
|Number of pages||16|
|Journal||Computational Linguistics in the Netherlands Journal|
|Publication status||Published - 18 Dec 2018|