英语论文网

providing subscores for fluency, pronunciation
and grammatical accuracy as did the original TSE, the revised TSE
now reports scores on a scale of communicative language ability. This
new scale takes into account the effectiveness of communication
resulting from a number of linguistic, sociolinguistic, discourse and
strategic competencies.
Because context is central to the construct of communicative competence,
the revised TSE uses tasks that are specified in terms of
seven contextual characteristics: test interviewer, audience, setting,
topic, purpose, function and visuals. The purpose of these features is
to provide appropriate and rich sociolinguistic and discourse features,
in terms of task characteristics, and to enable examinees to engage
their communicative language ability in responding to test tasks
(Henning et al., 1995). As a result of these changes, the revision of
the TSE is now held by Educational Testing Service (ETS) to be:
a test of speaking ability designed to evaluate the oral language proficiency ofnon-native speakers of English who are at or beyond the postsecondary levelof education . . . [When] used in conjunction with other measures, it can helpprovide an indication of the examinee’s ability to successfully communicate
in English in an academic or professional setting.
The construct underlying the revised TSE is said to be the ‘ability to
accomplish specific language tasks comprehensibly, accurately,
coherently, and appropriately with respect to specific
interlocutor/audience, topic, and purpose’ (Educational Testing Service,
1994: 1).This conception is more or less consistent with thoughtful explicationsof communicative competence that have been provided by,
D.E. Powers, M.A. Schedl, S. Wilson Leung and F.A. Butler 401
among others, Bachman (1990), Canale and Swain (1980), Chapelle
et al. (1997), Douglas and Smith (1997), Duran et al. (1985), Henning
and Cascallar (1992) and Stansfield (1986), each of whom has
contributed to a better grasp of the concept. However, as Douglas and
Smith (1997) have suggested, even though the term ‘communicative
competence’ has been used for three decades, the concept is still not
well understood.The study described here focuses on what we have termed ‘communicativesuccess’, defined as the ability of listeners to respond correctly,appropriately or positively to a speaker’s message. Our interestwas in determining the degree to which examinee performance onthe TSE is related to the communicative success of those who listen
to them.
1 Previous research on TSE
In their study of a prototype version of the revised TSE, Henning et
al. (1995) found the correlation between scores from the old and the
new versions of the test to be .83, suggesting that both versions tap
similar constructs. (This high correlation does not, however, tell us
what this construct is.) Moreover, the investigators found that performance
on both the old and the new versions of the measure correlated
strongly (r = .75 and r = .82, respectively) with an oral language
interview which, according to the authors, is recognized by the
Foreign Service Institute, the Interagency Roundtable and the American
Council on the Teaching of Foreign Languages (ACTFL). This
evidence is also consistent with the view that the two versions are
measuring constructs that are very highly related.
In light of the