Cambridge Exam Updates

Last Saturday morning I attended a seminar on Cambridge exams and this is a report on the most significant points covered:

Cambridge offers “standard version” and “for school version” exams. The speaker made it clear that the introduction of “for school version” exams is the result of an informed decision and is not justified by the test-takers’ age.

The decision as to which exam best suits a test-taker’s profile should be taken after engaging in a critical analysis of Cambridge handbooks, which contain precise information.

The format, content and assessment criteria for both versions are identical.

The level has been preserved. What counts is the level accredited by the exam. The school version is not easier.

There’s a quality assurance system that is in charge of Cambridge exam validity.

KET exam deals with everyday English at a basic level.

Pet exam (preliminary): deals with everyday English at a pre-intermediate level.

Reliability is a key factor. A reliable test can be depended on to produce very similar results in repeated uses.

Reliability is guaranteed by a comprehensive description of the levels.

The exam validity can be classified into:

A. Construct validity: resides in the interaction of the underlying cognitive ability and the context of use. In the past it was based on the CLA model: communicative language ability, designed by Backman in the 90´s. It has been replaced by the socio-cognitive framework, designed by Professor Cyril Weir in 2005. The latter offers the distinction between an internal mental process (cognitive) of the test-takers and the external contextual factors (social). The standard and the school versions have the same level (cognitive) but the context is different.

B. Content validity: is connected to which extent the content of the test represents the target language domain; the area of interest. It is based on the schema theory: the mental structure that represents some aspects of the world. It depends on a collection of organized, interrelated ideas or concepts: the knowledge of the world. It's connected to the reference store from which a person can retrieve relevant existing knowledge (experiential background). One retrieves/remembers information that is relevant to one's own schema. One will try to resist information that is not relevant to one’s own schema. When automatic thought is triggered, one acts effectively without effort. Accessibility is defined as a cognitive shortcut. There is a direct impact on the way the individual processes the task set up in the context.

C. Scoring validity: implies what has to be measured. In the case of writing there is a General Mark Scheme:

Band 5: All content elements covered appropriately. Message clearly communicated to reader. (Excellent)

Band 4: All content elements adequately dealt with. Message communicated successfully, on the whole. (Very good)

Band 3: All content elements attempted. Message requires some effort by the reader. Or, one content element omitted but others clearly communicated. (Good)

Band 2: Two content elements omitted, or unsuccessfully dealt with. Message only partly communicated to reader. Or, script may be slightly short (20-25 words). (Fair)

Band 1: Little relevant content and/or message requires excessive effort by the reader, or short. (10/19 words). (Poor)

Band 0: Totally irrelevant or totally incomprehensible or too short (under 10 words). (Below standard)

In the case of speaking, it implies the interplay between the underlying cognitive ability and the context of use. Now they’ve introduced the terms “speaking examiners” as they assess a spoken test.

In the case of listening, there is overlapping between context and cognitive validity. It’s ephemeral and it’s marked by the absence of paralinguistic information in a testing context.

I think it is necessary for ESL teachers to be well-informed so as to be able to help those students interested in having international certifications.