Homepage » Services » The KPG English Corpus

The KPG English Corpus

This Corpus comprises collections of written texts (scripts) produced by EFL speakers/learners in Greece, who have sat for the national foreign language examinations (known as KPG) in English. It has been developed over a period of seven years and, to date, totals to more than 3.5 million words. The Corpus collections are stored in a Script Database which is being enriched on an ongoing basis. The database maintains a balance in the representation of proficiency levels (A1-C2), types of writing tasks, types of texts candidates are asked to produce, and communicative contexts to which they are asked to respond using the foreign language. The scripts have been marked in accordance with the KPG grading grid specifying three performance zones (grading bands) across proficiency levels: Fully satisfactory, Moderately satisfactory, Unsatisfactory.

Researchers interested in obtaining a chunk of this data are encouraged to contact us at rcel[at]enl.uoa[dot]gr, specifying the amount of data they need, the research purposes for which they need it, how they will treat it, etc. The specific conditions for using the KPG data will be stated in an agreement, to be signed by the RCeL Director and the researcher.