Collection of  written productions of L2 (English) students, annotated in CEFR by experts and containing additional metadata. Data was collected in two French universities between 2018 and 2023. Students are not following a curriculum dedicated to the English language, but students from other curriculums learning English to communicate in their specific domain. 

Students had a choice between three prompts.
Task 1 : Décrivez et donner votre opinion sur une oeuvre d'art, expérience scientifique, découverte, invention, technologie, technique, concept ou théorie de votre champ disciplinaire (describe and give your opinion on a work of art, scientific experiment, discovery, invention, technology, technique, concept or theory of your domain)
Task 2 : Expliquer la source d'inspiration pour le choix de vos études/métier (une personne ou situation qui vous a inspiré)  (explain your inspiration for the choice of your studies or job, a person or a situation that inspired you)
Task 3 : Faire une demande de bourse pour un projet personnel lié à vos études (séjour à l'étranger, année de cesure, stage, recherche...)  (make a grant enquiry for a personal project revolving around your studies)

Dataset is a comma-separated file with the following columns :
 - Date_ajout : date at which text was written and collected
 - pseudo : unique anonymous identifier for a student
 - Voc_range : CEFR level corresponding to the range of vocabulary used in the text, as given by expert teachers 
 - CECRL : CEFR level of the written text, as given by expert teachers
 - nb_annees_L2 : number of years spent studying the language, self-reported
 - L1: first language/native language/mother tongue of the student, self-reported
 - Domaine_de_specialite : curriculum/domain of study followed by the student at university, self-reported
 - Sejours_duree_semaines : number of weeks spent in an English-speaking country, self-reported
 - Sejours_frequence : number of trips taken in an Englihs-speaking country, self-reported
 - Lang_exposition: number of hours spent exposed to the language (television, music, radio, etc.)
 - Lecture_regularite : how often the student reads in English, as a qualifier ("jamais"=never, "mensuelle"=monthly, "hebdomadaire"=weekly, "quotidienne"=daily), self-reported
 - autre_langue : which other languages are spoken by the student if any, self-reported
- tache_ecrit : choice of written task between one of the three prompts
- Texte_etudiant : student text, cleaned from HTML tags and with names anonymized
- Section_renforcee : whether the student followed an advanced English section in high school or earlier at university

celvasp_full_annotated_with_metadata_2018_2023.csv

Site en cours de construction.&nbsp;
Merci de vous reporter à l'article suivant:&nbsp;
&nbsp;
Mallart, C., Simpkin, A., Venant, R., Ballier, N., Stearns, B., Li, J. Y., &amp; Gaillat, T. (2023). A new learner language data set for the study of English for Specific Purposes at university level. Proceedings of the 4th Conference on Language, Data and Knowledge - LDK 2023, 1, 281–287. <a href="https://hal.science/hal-04247635" target="_self">https://hal.science/hal-04247635</a>&nbsp;

Corpus d'Etude des Langues Vivantes Appliquées à une Spécialité

Mots-clés	NLP Learner data Second language
Auteur :	Thomas GAILLAT
titre	Celva.Sp_2018-2023
http://nakala.fr/terms#created
licence	CC-BY-4.0
type	http://purl.org/coar/resource_type/c_ddb1
mots-clésen	NLP
mots-clésen	Learner data
mots-clésen	Second language
descriptionfr	Set de données d'apprenants d'anglais, contenant les textes écrits par les apprenants avec leur niveau CEFR tel que donné par des annotatrices expertes, ainsi que des métadonnées sur l'apprenant.
descriptionen	English leaner dataset, containing texts written by French students of English as a second language. This dataset also includes the CEFR level of the written text as given by expert annotators, as well as metadata on the student.
langues	en
langues	fr
auteur	Thomas GAILLAT

Corpus d'Etude des Langues Vivantes Appliquées à une Spécialité

Celva.Sp_2018-2023

Données

Visualisation