Word segmentation in texts written by first year primary school children: a study in the Spanish context

Abstract

INTRODUCTION. One of the basic aspects for the acquisition of adequate written communicative competence in Primary Education is the knowledge and use of spelling and, in particular, word segmentation. Thus, within the framework of the Scolinter research project carried out in collaboration with the University of Grenoble Alpes and the University of Milan Bicocca, this study has been developed at the University of Almeria in order to analyse the segmentation of texts written by first-year students of Primary Education from the Spanish context. METHOD. This study has been carried out according to the following methodological procedure: collection of texts written from the visualization of four images of a narrative sequence; design of the Scolinter corpus made up of written texts from the collection, transcription, annotation and automatic treatment of the language using the computer tool Aliscol; description and analysis of texts in the Spanish context on word segmentation. RESULTS. Among the most relevant results, it was found that 74.7% of the written texts present cases of unconventional segmentation. According to the typology of these, there is a predominance of cases of hyposegmentation compared to hypersegmentation, the most common being those in which the first formant involved is grammatical and the second is a lexical formant. DISCUSSION. In agreement with previous research, we found hyposegmentation represents the most frequent type of unconventional segmentation in texts written by first-year students of Primary Education. This study has contributed not only to the advancement of educational research from a psychogenetic perspective, but also to reflection and improvement on the teaching of writing from an approach aimed at understanding the reasons why students perform unconventional segmentation.