Testing Reading Comprehension

By | February 25, 2021
Testing Reading Comprehension


When we come to the testing of reading comprehension, we are concerned with the testing of students who have passed beyond the purely audio-lingual state of language learning, in which “reading” and “writing” are used simply to reinforce the oral/aural learnings, and have proceeded to a stage in which reading and writing are taught as skills (or, more properly, as complexes of skills) recognized as useful in themselves.

Because of the importance of this point, let us amplify it a bit. In most modern foreign language courses, instruction begins with the teaching of the sound system and the most frequent and/or useful grammatical patterns of the spoken language. Vocabulary is at first quite limited, but as the student gains control over the sounds and structures of the language, he is “fed” more and more vocabulary, chosen for its usefulness in oral communication. During this stage, such reading as is presented to the student is designed primarily to strengthen his control of the oral/aural skills. Readings are generally built around the grammatical structures and lexical items which the student will need to communicate orally, and the content of the material is regarded as relatively unimportant. When the teacher considers that his class has gained good functional control of the spoken language, he may introduce reading as an end in itself. At this point the class enters an area where some students will undoubtedly advance much faster and much further than others, for reading involves, for both native speaker and foreign learner, the manipulation of a complex of skills only part of which is, strictly speaking, linguistic. The abilities needed in reading a language include at least the following:

1. Language and graphic symbols

a. Comprehending a large percentage of the lexical items occur-ring in nonspecialized writing and being able to derive the meaning of unfamiliar items (or special uses of common items) from the contexts in which they occur

b. Understanding the syntactical patterns and morphological forms characteristic of the written language and following the longer and more involved stretches of language (sentences and sequences of sentences) occurring informal writing

c. Responding correctly to the graphic symbols of writing (e.g., punctuation, capitalization, paragraphing, italicizing) used to convey and clarify the meaning

2. Ideas

a. Identifying the writer’s purpose and central idea

b. Understanding the subordinate ideas which support the thesis

c. Drawing correct conclusions and valid inferences from what is given

3. Tone and style

a. Recognizing the author’s attitude toward the subject and the reader; understanding the tone of the writing b. Identifying the methods and stylistic devices by which the author conveys his ideas In practice, the above abilities are mutually dependent. A writer may, for example, elect to use humor to make the reader aware of some common human failing, conveying the ludicrousness of this human behavior by an unusual selection of lexical items. The good reader, then, is one who can respond simultaneously and appropriately to the language, ideas, and stylistics of mature writing, and, moreover, can achieve these understandings with reasonable speed and fluency.’ It is the foreign student’s acquisition of such a complex of abilities that we wish to measure in reading comprehension tests.


It has been found that the same general type of test long used to measure the reading ability of native speakers of English will work with equal effectiveness with foreign learners of the language. Such a test consists of a number of short passages of varying styles and content, each followed by a series of multiple-choice comprehension items. By a judicious selection of his passages and a careful working of his items, the test writer is able to test the examinee’s understanding not only of the surface meaning of a passage but also of the author’s purpose, attitude, and method in fact, all the abilities listed in the previous section. Thus the reading comprehension test is somewhat parallel to the advanced-level auditory comprehension test in which the student is given rather long stretches of the oral language, perhaps in the form of dialogues, from which he must sift out and interpret a multiplicity of phonological, grammatical, and lexical signals occurring simultaneously.


  1. Length. Inasmuch as the test writer will generally wish to include samples of various kinds of material, the individual test passages should be kept brief. On the other hand, there should be sufficient content to yield at least six or seven comprehension items for pretesting. Passages of between 100 and 250 words are about the proper length. The specific purpose of the test will naturally dictate the subject matter of the passages selected. In a general screening test for foreign applicants to American universities, for instance, the selections should reflect the various kinds of reading material assigned in basic university courses. For these the test writer may draw upon such works as biographies, prose fiction, The matter of reading rate will be considered in a later section of this chapter.

2.These items may consist either of questions followed by several possible answers or of incomplete statements with several possible completions.

encyclopedia entries, and nontechnical articles on the natural and social sciences. The excerpts must be clear and meaningful when taken out of context, and not require outside subject-matter information to be fully comprehended. It cannot usually be assumed, for example, that the examinees will have the same kind of knowledge of American history, traditions, and cultural patterns as the American-born student. Nor should the subject matter be such as to give a marked advantage to students in particular fields (unless, of course, the test is designed for a special group of students such as science or education majors). On the other hand, the passages should not deal with information that is universally known, for, in this case, the candidates may be able to answer the questions correctly without paying much attention to the passages. (Thus a paragraph dealing in a general way with a widely known historical event or personage would constitute an unreliable test of reading comprehension.)

3. Style and treatment of the subject. As already suggested, reading tests should generally include materials of various types and styles, though all should possess reasonable merit as pieces of writing. The test writer will soon learn, however, that the literary excellence of a paragraph is not in itself a guarantee of its suitability as a test passage. Paragraphs that make just one clear, direct point, for instance, seldom make suitable passages for testing purposes, inasmuch as they do not yield a sufficient number of test items. Much more likely are paragraphs which (a) deal chronologically with a series of events, (b) compare or contrast two or more people, objects, or events, (c) present an author’s individualistic opinions on a familiar subject.

4. Language. Although, as stated above, the passages in a reading test should approximate materials which the examinees are likely to encounter when they put their English to use, the test writer must always set realistic tasks for his test population, which consists, after all, of learners of a second, or foreign, language.

Therefore, passages that are overloaded with extremely difficult lexical items and/or complex syntactical structures may have to be adapted somewhat to bring them within reach of at least the more proficient candidates. The simplification of reading passages must, however, be careful.

Leave a Reply

Your email address will not be published. Required fields are marked *