According to Miller’s book, he continues to reflect in this fourth edition that “learning is best fostered by encouraging students to read, reflect, and write about serious issues” (Miller p. vii). While this book takes processes such as this into account, generally speaking it is not open to a large variety of learning styles, mentalities, and other differences which may be just as well facilitate learning through other processes than in the traditional ones.
Undoubtedly, in any case, there are methods better than conventional testing which would better measure a students learning. This could be done in a multitude of ways, while the most effective seem to be comprehensive open material tests, oral exams, and lengthy research-style reports.
Conventional testing has proved to be a poor measure of ability for a number of reasons. First, it only shows if students have learned a small percentage of specific information which is supposed to be comprehensive and specific enough to reveal if a student has learned the majority of the information taught. Here the flawed assumption is made that if a student has learned a certain percentage of the material in the exam, then they thus have learned the same percentage of material taught over the section which is being tested.
While this may be partially true to some extent, it is certainly not wholly true and should not be assumed to be true on average. More comprehensive tests could reveal a larger amount of learning, while materials being allowed for these kinds of examinations would not be as harmful as some speculate if the tests are designed properly. Conventional testing does not commonly allow for materials to be used in the majority of cases, and even when this material is allowed the conditions which do allow for the misuse of material or even student collaboration exist.
The most effective amendments to the conventional test are thus fully comprehensive tests which measure the clear majority of subjects taught during the section tested, the use of material in such a way that the material facilitates research rather than provides direct answers which could be obtained in the given time frame without prior understanding of the material, and ensuring this level of comprehensive examination can occur within a reasonable time frame so that it can be supervised.
If these were the elements of conventional testing, the major arguments against its ability to measure and major flaws would not be as strong, and thus conventional testing would not be the subject of increasing scrutiny. Conventional testing has been considered nearly counter-productive for these reasons while they create a sort of forced learning style (Popham p.136).
Oral exams also facilitate a better measurement of learning. Though more time consuming, they provide a direct demonstration of any number of concepts learned, while an instructor has the ability to better see potential flaws in knowledge while attempting to exploit them. As such these tests have the potential to reveal with relative ease what a student has not learned as well as what they have learned. Other than the challenges involved in equal grading this method would undoubtedly provide a great deal of additional information of what the student has or has not learned and thus would be a better measure than conventional testing.
Lengthy term papers assigned as a measure of knowledge are often employed in many classes, though they are not as common across a large number of subject areas as they could be and thus are not commonly regarded as conventional testing. It is possible to employ these kinds of assignments to measure ability in areas of math, physics, anatomy, and other areas which they are not commonly used.
Term papers have the advantage over tradition testing in that they are more comprehensive while revealing more information what the student has learned, and while this is similar to the fairly common “take-home test” in this regard, term papers have the ability to scanned and checked for cheating whereas take-home tests are not checked as easily.
There are undoubtedly more methods to better measure learning that are not included in the ideas discussed. Simply put, the majority of current conventional testing methods only partially test what a student has learned and almost never measure what they have not learned, all the while making an unfair proportionality assumption and using this to rate the student. Clearly this method can use serious improvements, and hopefully for the sake of students or anyone that cares about a more accurate measurement, the method is improved while becoming commonplace.
Works cited
Miller, Robert, The Informed Argument: A Multidisciplinary Reader and Guide, Harcourt Brace College Publishers, 1995.
Popham, W. James, Test better, teach better: the instructional role of assessment, ASCD, 2003.