Testing

The CERCLU project has been carefully and deliberately staged in order to achieve

• a constant typology of valid test-input (texts and related comprehension questions)

• rapid administration

• a limited-in-time testing process,

• computer assisted assessment,

• automatic marking,

• automatic interpretation of performance.

English and Italian are the languages selected.

The choice of the unmarked language use of everyday communication seems inappropriate, or at least insufficient, for our purposes, while the marked language related to macro-areas of study introduces the problem of identifying these areas and the related specific communicative language uses, i.e. contexts and functions which are both frequent and relevant to appropriate everyday interaction within the academic world.

For both levels (B1 and B2) we keep the operative distinction into subtests linked to the four traditional abilities, and to differentiate between the relative authentic texts. The differences between our B1 and B2 level tests do not only regard surface aspects such as the length of the texts, the number of alternatives for the answers, and the index of frequency of words. We want the complexity of the elaboration process required of the candidates for giving their answers to be an important differentiating factor. Figure 1 shows this differentiation in the listening subtest section.  

e-mail