160 / 2018-03-15 09:34:41

The Construction of Computerized Dynamic Assessment（C-DA） Model for EFL Listening

Cognitive Diagnostic Attribute, Dynamic Mediation, English as Foreign Language Listening

全体主题 > Principles and methodology of developing standards for language assessment

Dynamic Assessment(DA) grounded in Vygotsky' s (1987) sociocultural theory can track how learners' potential abilities emerge and develop by offering graduated prompts as mediation when they encounter difficulties (Peohner, 2015). While Cognitive Diagnostic Assessment(CDA) distinguishes itself in identifying learners' cognitive strengths and weaknesses with finer-grained attributes such as knowledge structures and processing skills underlying learners' response scores. The combination of the two hold a great potential in promoting students' language ability. Therefore, this study integrating DA and CDA extends traditional“assessment of learning” to “assessment as learning", so as to achieve a fine-grained EFL listening diagnosis and provide effective mediations for EFL learners
Through literature review, verbal protocols and experts' coding etc, the study initially established an attribute-based mediation model for EFL listening which was between the interactive and interventional form. Then the model was used to conduct three one-to-one dynamic assessment experiments on 8 low-middle level freshmen once a week. The results indicated that: 1) After three interventions, the listening scores of the experimental group were significantly higher than those of the control group (Test 1: t = 1.144, p = 0.258, effect size = 0.451; Test 2: t = 2.224, p= 0.031, effect size = 0.863; Test 3: t = 3.402, p= 0.001, effect size = 1.49). In addition, with the increase of test difficulty, the actual score and mediated score of students in the experimental group also showed significant differences, indicating that regardless of their independent performance, participants generally benefited from mediation,which indeed explored the Zone of proximal development of the group. 2) Through interviews, six of the students thought that the attribute-based mediation was very effective and brought assessment and instruction into a unified activity, which was embodied in diagnosing students' current level, correcting listening problems in time, and enhancing the awareness of using listening strategies, especially note-taking and previewing items. However, when students were overwhelmed by too much mediations, they were more easily to have negative emotions, resulting in a lower degree of acceptance for the mediation. By applying the cognitive diagnostic listening test as a powerful diagnostic tool, this study only provided a reference template for the implementation of computerized dynamic mediation, but the effect of its large-scale use and instruction values need to be further explored.