Back to search

National and Kapodistrian University of Athens

Γραμματικά λάθη των μαθητών της νέας ελληνικής ως δεύτερης γλώσσας και η κωδικοποίησή τους βάσει ηλεκτρονικών σωμάτων κειμένων

Abstract

dc:description

The thesis offers an analysis of specific categories of grammaticality errors such as morphological and syntactic errors, which occur in short written essays produced by learners of Greek as a second language (GL2) during their exams of attainment in Greek for the B2 level. The study of these errors is based on the theory of interlanguage that views errors as evidence of learners’ lack of competence in L2 that occur systematically. Such shortcomings are considered as integral characteristics of the transitional linguistic system used by learners called interlanguage, which deviates both from their mother tongue and the target language, and is a natural and unavoidable stage of L2 acquisition.More specifically, a systematic collection, organization, classification and coding of GL2 learners’ errors is undertaken through the use of the methodological tools of Computer-aided Error Analysis. These tools are the development of a relatively representative error-tagged Computer Learner Corpus and an error taxonomy that classifies errors according to linguistic categories, as well as ways that learners modify the surface structure of the language. Thus, a quite extended learner corpus of L2 Greek was produced, the Corpus of Greek Learner Texts (CGLT), comprising approx. 76.000 words. The annotation of errors in this corpus was made with the “Episimiotis” software, so that it could be further available for research. Then a test research is conducted in order to evaluate the reliability and usefulness of CGLT, as well as of the error annotation system under use.Within the framework of the aforementioned test research 42 categories of morphosyntactic errors are quantitatively analyzed by means of their relative frequency and the error rate of several linguistic features related to them. This analysis is conducted in the entire CGLT as well as its subcorpora, divided according to the learners’ mother tongues and the genres of written essays, in order to verify whether these variables have an impact on error categories or not. Subsequently, the results of this analysis are investigated in order to verify the curriculum predictions for levels B1 and B2, as regards GL2 learners’ knowledge of 16 different morphosyntactic features with discriminative weight for these levels of attainment. The main aim of this correlation is highlighting through empirical data the boundaries of level B2, as well as the descriptors of the related curriculum that need improvement.In addition, the 10 most frequent and the 10 most rare errors in CGLT are submitted through a questionnaire to a team of instructors of GL2 in order to evaluate their gravity in terms of linguistic knowledge, communication and frequency. The aim here is to identify the instructors’ conceptions concerning morphosyntactic features of GL2 that need more attention in class.The most persistent errors in the interlanguage of level B2 learners, as indicated by the research results, concern the use of articles, verb aspect, prepositions (in adverbials, verb complements or adverb complements), complementizers, verb complement case and modality particles (θα, να, ας), the lack of grammatical gender agreement between article or pronoun (except clitics) or adjective and noun, the choice of grammatical gender and the errors in clitics’ syntax (mostly deletion). Among the previous errors only the incorrect use of articles and the wrong choice of grammatical gender were found to be related partially with learners’ L1 transfer, while text genre does not seem to have an significant impact on learners’ interlanguage. Among the investigated morphosyntactic features mentioned in the curriculum for levels B1 and B2 the most error-prone one was found to be the combination of the verbs αρέσω (like), ενδιαφέρω (interest), ενοχλώ (annoy), πάω (go) and αναστατώνω (disrupt) with their subject, the morphology of neutral nouns ending in -ος/-ους, the morphology of ancient-Greek-origin adjectives ending in -ης, -ης, -ες and the syntax of verb αρέσω (like) complements (mostly clitics). Finally, instructors’ conceptions showed a quite different view, as they point to the incorrect use of nominal case, grammatical gender and verb complement, lack of grammatical gender and number agreement, incorrect syntax of comparative clauses and the incorrect use of the negative particles δεν and μην as serious errors. Instead, instructors regarded as non-serious errors those related with the use of articles, verb aspect and prepositions contrary to the findings based on CGLT data.In conclusion, some proposals based on previous findings are made for the improvement of the curriculum predictions of level B1 and especially B2 as regards frequently erroneous morphosyntactic features, as well as the means of focusing on them effectively during instruction. Moreover, it is pointed out that the basic findings of this research agree in general terms with the conclusions of other similar studies. Finally, it was found that CGLT can shed light on how learners tend to use the most frequent and basic linguistic features as well as on the patterns of related errors, but cannot give adequate evidence as regards the use of more rare or more marginal linguistic features, the more focused analysis of multi-word features with specific meanings or the estimation of under-use (avoidance) of difficult grammatical types. To amend these shortcomings in the future the expansion of the CGLT and the development of a comparable corpus of native speakers’ written texts are proposed.

Degree

thesis:*
Grantor dc:publisher
National and Kapodistrian University of Athens
Year dc:date
2018

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Tzimokas, Dimitrios
  • Τζιμώκας, Δημήτριος

Subjects

dc:subject × 40

Rights

dc:rights
Statement dc:rights
  • BY
Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/44648

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Tzimokas, Dimitrios; Τζιμώκας, Δημήτριος. Γραμματικά λάθη των μαθητών της νέας ελληνικής ως δεύτερης γλώσσας και η κωδικοποίησή τους βάσει ηλεκτρονικών σωμάτων κειμένων. National and Kapodistrian University of Athens, 2018. http://hdl.handle.net/10442/hedi/44648