Measured Constructs: A History of Cambridge English Language Examinations 1913-2012
This hefty, academic tome, which is No. 37 in the series Studies in Language Testing, is by Cyril Weir, Director of the Centre for Research in English Language Learning and Assessment (CRELLA). With his two co-authors, he has produced a comprehensive history of the Cambridge exams from their birth in 1913 down to the present day – quite a daunting task! As the title suggests, the book is aimed at an academic audience and presupposes some knowledge of linguistic terminology. However, since Weir writes in a clear, interesting way, the book could also appeal to teachers who are interested in the Cambridge English Language exams and how they have developed over the years. In fact, one of the most interesting aspects of the book is the way in which it shows how pedagogical theories and teaching methods, as well as socioeconomic factors, have affected the content and format of the exams, especially over the last 40 years.
Weir believes that a knowledge of the past helps us to understand present-day theories and methods in the field of language teaching and testing. To emphasize this point, he starts his book with a (non-translated!) quotation from Cicero: Nescire autem quid antequam natus sis acciderit, id est semper esse puerum, which I freely translate as ‘If you don’t know what happened before you were born, then you’ll remain an ignoramus all your life’. Many teachers would agree with this statement, especially since there’s a tendency in EFL circles to assume that modern theories and methods are automatically superior to previous ones, often forgetting that many former methods of teaching and testing also had their virtues, as well as their weaknesses.
The first Cambridge exam was the CPE, which was taken in London in 1913 by three candidates. Last year over four million people in 130 countries took the various Cambridge exams. Quite a development! The first exam, which was largely aimed at foreign (mainly French and German) teachers of English, consisted of an oral part (dictation, reading aloud, conversation) and a written part (translation, grammar, English essay, English literature, phonetics). It lasted 12 hours (!), compared to under five hours today. As time passed, various parts of the exam were dropped, e.g. phonetics (1932), dictation (1975), translation (1975), literature (1975), and reading aloud (1986). Generally speaking, from the 1970s onward, with the spread of the communicative approach, there was a movement away from the more academic-type of exam, with its literary slant, towards a more language-focused format.
One significant development in the field of testing was the rise of the psychometric, multiple-choice type of exercise (which originated in the USA) before and after World War I, mainly because of a need for a quick, reliable way of testing the huge number of immigrants entering the country. This type of exercise was only adopted at Cambridge after The Cambridge-TOEFL Comparability Study in the late 1980s, which compared the FCE and TOEFL exam formats. The Cambridge authorities then realised that the inclusion of multiple-choice exercises could increase reliability and simplify marking. There was also money involved, as Weir mentions. The authorities, who were becoming more professional (and money-minded) in the 1990s, feared that their TOEFL rivals would introduce their exams into Europe, so to forestall any such competition, they decided to include their own multiple-choice exercises. An example of how economic factors can influence exam content!
More recently, the pressure from the EU in the 1990s for a more standardized system of language assessment and the introduction in 2001 of The Common European Framework of Reference for Languages: Learning, Teaching, Assessment (CEFR) has led to the development of more precise descriptions of language ability, i.e. the assessment scales and descriptors with which we are familiar today. The introduction of double marking in written exams and the use of the Interlocutor/Assessor system in the oral exam are also attempts at achieving a more accurate assessment of language ability. Under this system the Interlocutor gives a holistic impression mark and the Assessor an analytic mark based on scales. The two examiners do not confer, and the result is the average of the two marks. Weir concedes that both approaches have their merits and weaknesses, and it is interesting that Cambridge uses this dual system.
To sum up: this is an academic, informative account of how a tiny examination with three candidates became the enormous, international enterprise it is today. If you have time and energy, it’s available from our marvellous ETAS Library!