| Budget Amount *help |
¥1,100,000 (Direct Cost: ¥1,100,000)
Fiscal Year 1996: ¥300,000 (Direct Cost: ¥300,000)
Fiscal Year 1995: ¥800,000 (Direct Cost: ¥800,000)
|
|---|
| Research Abstract |
With the goal of making a construct valid test of speaking ability in mind, this paper examines the components of speaking ability using the Multi-Trait Multi-Method analysis. The focus is on the analysis of the relationships among nine traits (pronunciation, grammar, fluency, discourse, content, vocabulary, interactional competence, sociolinguistic competence, and comprehensibility) and three methods (writing test, interview. test, and tape-mediated speaking test).
Three factors obtained from the factor analysis corresponds with three methods designed for measuring the ability. These three factors were named Direct Oral Communication Ability, Semi-Direct Oral Communication Ability, and Written Communication Ability individually according to the nature of each ability.
Nine traits in Direct Oral Communication Ability (Interview test) contribute to become the fundamental elements of the factor while they still maintain their own characteristics which cannot be replaced by other traits. This idiosyncrasy was made clear through the inter-trait correlation coefficients. The same is true for the nine traits in Semi-Direct Oral Communication Ability. The nine traits function as factor construct elements but they still have their distinctiveness.
The traits in each factor, even though they have the same naming (grammar, vocabulary, discourse etc), are working separately in the three methods.
Interview tests could measure part of the whole communication ability, and tape tests could measure another part of the whole communication ability. The traits in each method can be a good composite of the ability the method is measuring.
For future research, Computerized Adaptive Speaking Tests should be taken into consideration.
|
