فهرست مطالب

Language Testing - Volume:7 Issue: 2, Oct 2017

International Journal of Language Testing
Volume:7 Issue: 2, Oct 2017

  • تاریخ انتشار: 1396/10/30
  • تعداد عناوین: 5
|
  • Zahra Javidanmehr *, Mohammad Reza Anani Sarab Page 1
    Cognitive Diagnostic Assessment (CDA) is a type of educational assessment that is designed to measure specific knowledge structures and processing skills in students so as to provide information about their cognitive strengths and weaknesses (Leighton & Gierl, 2007). CDA has been instrumental in turning the attention of practitioners to more diagnostic, descriptive, and fine-grained levels of feedback. Different statistical, psychometric models, called Cognitive Diagnostic Models (CDMs), have been proposed to extract this kind of information from both diagnostically and non-diagnostically designed tests. These models provide two sets of information to the test users: information on mastery/non-mastery patterns of sub-skills for testtakers and information on the diagnostic power of test items. Due to its novelty and relative complexity of its procedures, cognitive diagnostic assessment is still far from achieving its proper place in educational assessment. This paper aims at providing an easy-to-grasp account of CDA's theoretical foundation and its procedures of test analysis. The present paper first focuses on what and why of CDA in education and second language acquisition. In this part, theoretical underpinnings of CDA, psychometric models of practicing the analyses, model selection, and studies in SLA are presented. The second section presents how these foundations are put into practice in a stepwise manner. Four main steps are delineated in conducting a CDA analysis. The procedural steps are then exemplified using real data for analysis. The paper concludes with an account of the limitations and untapped areas in CDA.
    Keywords: cognitive diagnostic assessment, educational measurement, high stakes testing, Qmatrix, reading comprehension
  • Shiela Kheirzadeh *, S. Susan Marandi, Mansoor Tavakoli Page 2
    The investigation of test-taking strategies is a crucial aspect of test validation studies especially if the test is a high-stakes one. The purpose of this study was to uncover if gender, academic discipline and the total exam score influence on the test-taking strategies employed by the candidates of the General English Section of the National Iranian PhD Entrance Exam which is a high-stakes test in Iran. Four sub-components of the strategies, namely, cognitive/metacognitive, test wiseness, time management and emotional strategies were investigated. The MANOVA results indicated no statistically significant difference between male and female exam candidates in the employment of strategies. However, Engineering and Basic Sciences academic discipline candidates outperformed the Humanities in the application of cognitive/metacognitive and test wiseness strategies. Moreover, the total exam score was influenced by the emotional strategies. The findings of this pioneering study also have implications for test designers, curriculum developers and teachers.
    Keywords: test-taking strategy, academic discipline, gender, PhD Entrance Exam
  • Gholam-Reza Kiany *, Parvaneh Shayestefar, Yasser Amoosi Page 3
    Teacher evaluation and certification is a proper approach taken to assess teacher knowledge base and to guarantee that teacher candidates are qualified and have met particular teaching standards. To this end, teacher-education university (i.e., Farhangian University) of Iran recently adopted a teacher evaluation and certification policy, nationally called ASLAH, which mandates all teacher candidates to earn a teaching certification by passing a comprehensive exam. The purpose of the present mixed-method study was to describe the development and validation of ‘a written assessment framework’, as one of the requirements of ASLAH project, that would measure English teacher candidates’ professional knowledge and competency to teach English. The data were collected through a series of interviews with teacher educators and subject-matter experts (N=15), questionnaire surveys that involved 320 English language teachers, and test performance of 62 English teacher candidates. First, the study dealt with the development of the hypothesized framework that included items assessing teacher Content-Knowledge (CK) competency by virtue of nine domain-specific courses recommended by experts’ qualitative and quantitative data. The items were empirically determined for content specification and validity, item difficulty and item discrimination. An initial piloting of the newly developed tool to teacher candidates showed tests as valid and reliable instruments for measuring teachers’ CK competency. Perceived ‘fairness’, ‘consequences’, and ‘quality of the results’ of the present certification policy and practice were also explored from the eyes of all participants. The results showed that they were not negative about the possible test consequences and fairness. However they did not appear to be strongly positive about the practicality of teacher evaluation and certification project in the present context of teacher-education university. Policy recommendations and implications of the findings were discussed.
    Keywords: Teacher Evaluation, Teacher Content Knowledge, Certification Testing, TeacherEducation University, Professional Development
  • Mahmood Samaie, Saeedeh Mohammadi * Page 4
    Test impact, widely recognized as the influence of testing on learning and teaching, affects a set of stakeholders including test takers. This study defines the construct of test impact on test takers and describes the construction and validation of the scale of test impact on test takers (TITT). 410 participants having passed a language test in University Entrance Examination (UEE) were asked to answer the questionnaire containing 64 items. Exploratory factor analysis was applied in the study and yielded evidence for the expected five factor structure of the TITT scale, including the components of test results, test awareness, test experience, test importance, and test sociocognitive effects. The final TITT Scale and its subscales consisting of 56 items demonstrated an acceptable internal consistency and expected levels of stability of the responses across time. Cronbach’s alpha was .93 for the global TITT scale and between .78 and .88 for the five subscales. Implications are discussed and suggestions are provided for possible utilization and improvement of the scale, and future validity testing.
    Keywords: scale development, test impact, test impact on test takers (TITT), TITT scale
  • Faezeh Ghazizadeh *, Khali Motallebzadeh Page 5
    Assessment plays a crucial role in any educational milieu (OECD, 2003). The researchers sought to investigate the possible impact of diagnostic formative assessment on student's listening comprehension ability and self-regulation. To this end, 46 participants were selected from among 78 students. A listening test was given to the participants to measure their level of listening comprehension ability. After 14 sessions (7 weeks) of instructing listening materials and experiencing formative assessment principles, the same listening test was administered as the posttest. Gathered data were analyzed and the results revealed that students in the experimental group outperformed those in the control group regarding their performances in listening test. Moreover, the results proved that student's self-regulation increased while they practiced diagnostic formative assessment based on the results obtained from pretest and post test. The results of the present study may be useful for those teachers who practice language for the sake of communication and forming student's knowledge by diagnosing their weak and strong points.
    Keywords: Diagnostic Formative Assessment, Listening Comprehension Ability, Self-Regulation, Classroom Assessment