Syntactic Complexity and Subordination in Written Discourse of Speakers of Croatian as a Second and Foreign Language

This research focuses on the analysis of syntactic complexity and subordination in written discourse of speakers of Croatian as a second and foreign language. In order to describe syntactic complexity, the research takes into account traditional factors in measuring complexity: the number of T-units, the mean length of T-units and the percentage of error-free T-units. Subordination is calculated on the basis of the ratio of the number of produced subordinate clauses and the number of T-units. Furthermore, the paper presents qualitative data on the sentence type produced by the speakers, the proportion of the subtype in the total number of subordinate clauses and the proportion of the subtype in the total production. The obtained data is compared to recent similar studies conducted in the Croatian language and other languages. The research in this paper is based on the corpus of 25 written compositions on the assigned topic written by native speakers of Polish, German, Spanish, French, Hungarian, English, Russian/Ukraine whose level of Croatian is at B1 level according to CEFR. The results indicate that there are individual differences between the speakers and that the measures of syntactic complexity are good indicators of syntactic development of second and foreign language speakers. The examinees' syntactic development is in line with their expected language level.