Relações entre o construto e as características do teste escrito do EPPLE eletrônico: um processo de validação

• Ano de defesa: 2015
• Autor(a) principal: Anchieta, Priscila Petian [UNESP]
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual Paulista (Unesp)
• Palavras-chave em Português: Linguistica aplicada; Linguagem e línguas - Estudo e ensino; Professores de linguas - Formação; Comunicação em língua estrangeira; Competencia e desempenho (Linguistica); Avaliação de proficiencia; Applied linguistics
• Link de acesso: http://hdl.handle.net/11449/138438; http://www.athena.biblioteca.unesp.br/exlibris/bd/cathedra/07-04-2016/000863597.pdf

This thesis presents an investigation about the proficiency assessment of foreign language teachers that happens by means of proficiency tests. Our goal is to describe and improve the written test of EPPLE (Proficiency Examination for Foreign Language Teachers), by verifying aspects that involve its internal validity, including, for instance, the construct validity and the content validity. According to Fulcher and Davidson (2007), the construct validity is the basis of evidence for the interpretation or the use of a test. The content validity, on the other hand, represents the group of tasks of which a given test should be a sample. To fulfill our goal, we present a detailed survey about the proficiency assessment theory that underpins the construct for the EPPLE examination, based on its tasks and goals. In addition to the considerations about the exam as a whole, we also present, as a focus, the tasks of the written test in order to contribute with investigations about its content validity. We look into aspects such as the test method, the tasks, the reliability and the validity that are involved in the implementation and improvement of an assessment instrument. Thus, by means of a theoretical discussion associated with the task analyses of the written test, the data collected in the application of EPPLE in 2012 and data from a research questionnaire, we present criticisms and suggestions that may contribute with the improvement of EPPLE. We indicate that there are many open questions, more indicated to test written production, in the section in which the assessment goal is reading skills. As for suggestions, for instance, we recommend that the EPPLE should better reflect its purpose of an exam for specific purposes, that is, the assessment of language that is representative of FL teachers' linguistic domain