apec-內頁

I. Conceptual Frameworks for setting and assessing English language standards

Two important social trends began to profoundly affect language testing at the end of the 1970’s and in the early 1980’s. First, waves of immigration, particularly to the Western world, meant that millions of people were using second and third languages in their places of work. It became obvious that some universal and objective standards for language ability were necessary if language skills were going to have effects as wide-ranging as safety in the job place to upward mobility in society. Second, national boundaries were breaking down, particularly in Europe, as people increasingly possessed the economic means to study, pursue careers, or even start businesses in foreign countries. Language testing was forced to become accountable to industry, government, and private individuals as it began to affect the vital interests of such players.

Assessment scales were able to meet the needs of the new global world in two ways. First, they provided a standard for defining and measuring language proficiency independent of specific languages and cultural contexts, opening up the possibility that the scores for tests of different languages to be anchored on the same set of standards (Bachman and Savignon, 380). Second, assessment scales developed alongside a new “proficiency movement” that was gaining strength at the beginning of the 1980’s. In the modern world, the “proficiency” proponents argued, it was no longer satisfactory to measure what learners knew about language, but to also measure what they could do with language. It excited language educators that through the new scales, language outcomes could be measured against a common metric, and that it was possible to accurately and objectively predict the degree of success with which an individual could handle language in a variety of situations (Omaggio, 330).

This section discusses four important and influential assessment scales, the “ACTFL Proficiency Guidelines” revised and published in the US in 1999, the “Common European Framework of Reference: Learning, Teaching, Assessment” (CEFR) published in 2001, the Canadian Language Benchmarks published in 2000 and the International Second Language Proficiency Ratings (ISLPR), first developed in 1978 as the Australian Second Language Proficiency Ratings, renamed the ISLPR in 1997. In addition to introducing the structure and unique features of each scale, we will discuss some of the inherent strengths and weaknesses in language scales.

The ACTFL Guidelines had a long gestation. During the Second World War, the United States faced unprecedented contact with foreign languages and cultures, and its military was forced to hastily set up programs in difficult languages like Japanese, Chinese, and Russian. The exhilarating successes of some of these programs combined with the realization that the US had not paid sufficient attention to language education in the first half of the 20th Century resulted in a preoccupation with foreign language through the 1950’s. The origins of the ACTFL Guidelines trace back to this period when, in 1952, the US Secretary of State Dean Acheson called on the Dean of the Language School of the Foreign Service Institute to create a set of criteria that could measure the foreign language proficiency of government employees (Stansfield).

In the following years, the Foreign Service Institute developed a radical new language assessment model. While previous language tests focused on achievement and the understanding of discrete points involving language, the FSI designed a simple 0-5 six-point scale ranging from no functional proficiency (0) to native-like proficiency (5), with each rank in the scale briefly and unambiguously defined in prose form. Furthermore, students did not face an examination under dry classroom conditions but took a face-to-face oral interview which simulated use of the target language under real-life conditions.

The FSI scale subsequently appeared in various incarnations and was adopted in many US government agencies, including the Peace Corps and even NATO. An Interagency Language Roundtable (ILR) finally appeared in 1985 which included full descriptions of the “plus” levels that had developed over the years. And an adaptation of the ILR designed for academic purposes, the ACTFL Proficiency Guidelines, were published in 1986 (Herzog, ILR). It is to the ACTFL Guidelines, revised most recently in 1999, to which we turn our attention now.

Because the ACTFL Guidelines were intended primarily for use in colleges and universities, the authors of the 1986 ACTFL Guidelines conflated the top three levels of the ILR, and created a broad category called the Superior Level. They further expanded the lower levels of the ILR (ILR 0-1), creating expanded sublevels of competency, which allowed the scale to target levels of ability that most adult learners attained, and that best matched the levels of learners of language for academic purposes. The 1999 revision broke with tradition and presented the guidelines in descending, rather than ascending order. The top-down offered the advantages of emphasizing that each level was closely related to the one found above it, and framed the language learning process in a more positive way. Furthermore, the full prose descriptions of each level were preceded by clearly delineated “thumbnail sketches,” which provided readers with a general understanding of the features of language used at each level and sub-level. The thumbnail sketches are provided in the chart below.