Essay Measuring

Measuring Language Skills

Whether we like it or not, in life it is hard to escape the phenomenon of evaluation. Sports competitions end with a final score, products and services are appraised, books, movies and other artistic productions are critiqued and rated, and inevitably, school performance is graded. When you enroll in a language class, one of the first things you are asked to do is to take a test to determine your ability level. During the term there may be numerous tests and quizzes and at the end of the course there will probably be a test to check what you have learned. Testing is used for purposes of placement, grading, and teacher/class evaluation and is an inevitable component of any learning program.

It is clear that placement testing is important for grouping together students with equivalent skills. The wider the skill range of a class, the more challenging it is for the teacher to address student needs and deficiencies and for students to remain attentive and engaged. Testing also provides the teacher with valuable information about what students know and what they need to work on. Throughout the course, testing supplies feedback to the teacher about how well teaching strategies are working and which activities are successful and which need to be altered. Testing at the end of a course not only provides information about a student’s proficiency and readiness to move up to the next level, but also serves as a means of evaluating teacher competency and effectiveness.

In order for assessment to be useful for all the purposes mentioned, the assessment instruments have to be well designed, properly used, and intelligently managed. Otherwise, assessment can be misleading and interfere with learning instead of promoting it.

These days there is a tremendous zeal for quantification. We crave precise, objective data free from bias and subjectivity. We want the facts, just the facts, please. We seek out the bottom line, that crucial, ultimate, number, which tells the whole story with unembellished, mathematical precision. We view the final grade as the decisive, last word about what has been accomplished and depend on standardized test scores to supply definitive information about proficiency. But perhaps, in something as complex and mysterious as the learning process, more is going on than can be communicated by a simple number. And perhaps the numbers we are given to look at mean less than we would like to believe.

Sometimes we can become so absorbed in quantification that we don’t pay enough attention to what the numbers measure and how well they measure it. We know that in everyday life, numbers are sometimes used to make subjective judgments appear unassailably objective. Judged Olympic events use measurements in the thousandths to try to project a sense of scientific accuracy where it doesn’t exist. Academic assessment, as well, tends to portray itself as more objective and precise than it really is.

In language testing and other kinds of academic testing as well, the quest for objective assessment often leads to testing elements (typically grammar and vocabulary in language classes) and using methods (such as multiple choice questions) that are amenable to quantification. The problem is that such elements and methods may not provide the best means of evaluating language proficiency. The things that are easiest to test and quantify are not necessarily the things most significant in language ability. In other words, there may be a discrepancy between test results and actual communicative ability, which is the ultimate goal of language instruction. Teachers, of course, see this phenomenon all the time. Students may do exceptionally well in tests on specific grammatical structures but persist in misusing those same structures in actual speech.

Unfortunately, high test scores rather than real linguistic ability may become the goal and measure of success of a class. If the ability to get high grades in tests of grammatical accuracy or knowledge of vocabulary becomes more important than the ability to comprehend, speak, read and write the language being studied, the whole learning process can be corrupted. The teacher’s efforts may shift from helping students become proficient language-users to making them more proficient test-takers, especially if the teacher’s own performance is measured by test grades.

Assessment, then, has the potential to be of great help or great harm in the language class. Proper assessment can greatly improve the quality and effectiveness of instruction while improper assessment can result not only in faulty evaluation but also in instruction that is not truly focused on learning. To assure proper assessment, the following ideas should be kept in mind: 1) testing is a tool and should always be subordinate to instruction rather than a dominant factor; 2) assessment that is humanistic and holistic is more reliable than mechanistic, narrow assessment; 3) assessment in language learning should focus on performance or the ability to use language rather than on knowledge about grammar, vocabulary or other language elements.

The dangers of allowing testing to be the master rather than the servant have already been discussed. Unfortunately, “teaching to the test” and gearing the class to the test rather the test to the class is quite common. When the tail wags the dog in this way, the quality of the class is diminished.

It is often thought that quantitative, mechanistic assessment is more accurate and reliable than humanistic, holistic assessment, which is viewed as fuzzy and subjective, but that is not necessarily true. For one thing, quantitative assessment typically measures specific components, and not the skill as a whole. Holistic assessment, on the other hand, is about the whole skill and not a specific, quantifiable element of it. Quantitative testing also assumes that the ability to correctly answer specific questions on a test is representative of overall linguistic ability. In fact it may just indicate that a student happens to know (or not know) those specific elements. The elements that are chosen for testing are themselves based on subjective determinations. So while identification of errors is objective, the criteria on which the whole assessment is based may be quite subjective. The “objective” mechanistic system may turn out to be, ironically, more subjective than the “subjective” holistic system.

Finally, in the area of language learning, the ultimate goal of the student is not to acquire information or knowledge but to actually be able to use the language. Testing that is limited to checking what the student knows instead of what the student is can do, falls short of the mark. Proper assessment of language skills entails assessment of real language use.

What all of this means for language schools, teachers and students, is that the best program is one which focuses on student learning and not on test scores. Programs which specialize in raising TOEFL scores may well achieve their goal, but they may not be successful in actually improving students’ language ability. The way to determine the quality of a language program is to ascertain its success in imparting usable language skills to its students.