| Popis: |
This chapter presents the English language tests which have been designed and validated in a longitudinal quantitative study in order to determine the effects of Content and Language Integrated Learning (CLIL) on English language competence. The sources used to formulate the tests, the steps followed for their design and the procedures employed for their validation are all expounded on. The ways in which their reliability, content, construct, and face validity have been guaranteed are outlined. Moreover, the statistical operations performed to guarantee them are showcased (Cronbach α, Kuder-Richardson reliability coefficient or item difficulty and discrimination). The actual tests are then presented in a format which can be directly applied in any CLIL classroom in order to determine the effects of CLIL on language competence. We are in dire need of empirically grounded data in this area and it can only ensue from the employment of empirically valid and reliable instruments such as those presented herein. The latter data will be crucial to fine-tune, reengineer or revamp CLIL implementation in order to keep it on track. |
