Creating and writing fair assessments is a difficult task. Developing tests that are both fair and accurate is a difficult and daunting practice for many. The creator must be unsure that the tests and assessments that are created are both reliable and valid. If they are not, the results will be skewed and a true and accurate assessment will not be conducted.
In order for an assessment to be considered to be reliable there needs to be a basis for the choice of questioning or other method of grading to be conducted. The methodology that is used needs to have been demonstrated by others in the field previously so that there is credibility in the selection and style of choice. In order for a test to be reliable, it has to be ensured that there are no mistakes, including anything that can be construed or interpreted as something else, even by accident. The information needs to be clearly asked and the ability to answer in a specific manner needs to be available. As such, the definition of terms and ideas need to be both clear and consistent. One manner in which this can be done when creating a test is to have a peer of the test creator proofread the test to ensure that no questions with possible double meanings have been used. This peer also can ensure that the material being questioned has been done so in a clear and concise manner. The goal is not to confuse the test-taker, but rather ask what information should be shared to help ensure that the examinee has every available opportunity to share an adequate representation of the knowledge that has been acquired.
Validity in the test-taking process is represented through the insurance that the questions being asked and the expected results are as clearly stated as possible. Some methods to ensure valid results when creating assessments include having information shared in a triangulate manner, using member checking strategies, and to ensure that there is no bias expressed in the questioning techniques. The triangulate manner is used to create questions from different sources to ensure the results are accurate. Member checking strategies is a system in which the creator of the material uses a second source to ensure the material being evaluated is accurate rather than relying on only one source for information and assuming that the one source is correct. Especially when open-ended questions are used, a test creator needs to ensure that there is no bias, or particular view, that is being more heavily favored than other views. In open-ended questions, the responder often will have a different view than the creator. As long as the responder can support the position stated with relative information, the difference in opinion should not be the focus, rather, the focus should be on the responder’s ability to justify the presented argument.
It is essential to create assessments which are both valid and reliable. An assessment that addresses both of these issues is more fair, gives the evaluator a better understanding of what material has been better understood by the responders and what material is not as clear, but also helps the evaluator realize, if there are many responses that are similar and wrong, that there is a particular area which needs to be revisited and perhaps have instruction given in another manner. If there is a particular question or questions that receive the same incorrect answer, the question or questions may be flawed in their presentation, may have an unintended double meaning, or the material may have been presented in an unclear manner. The question or questions need to be reconsidered and checked for both validity and reliability. If it cannot be substantiated that the questions are both reliable and valid, the questions should not be used (Creswell, 2005).
Pepper and Pathak (2008) describe a teacher assigning a grade for work as a challenging experience for both the teacher and the student. Teachers struggle because what they want the students to learn and remember is more important than the grade value assigned to the learning of the material. Teachers also realize that some students test better than others. The students with the highest scores do not always understand the material better or remember more of the details. Additionally, some of the students with the lowest scores may have learned more during a particular unit than peers in the class, but do not express that knowledge well during an assessment. Good teachers strive to try and create assessments that allow students the best possible way for the students to express their comprehension of the material within the unit or lesson. In that way, a good teacher is always evolving one’s craft in creating valid and reliable assessments (Pepper & Pathak, 2008).
The strategy of norming is a process by which results are taken from a large group and analyzed for trends. For instance, if 30 students take a multiple-choice history test and 25 of the students get number four wrong, there may be an issue. Through the use of norming, it may be discovered that 22 of these students selected “A” instead of “C.” Upon further evaluation it may be discovered that the question had bias or the two choices, if read differently, might have the same meaning or meanings that are too similar to determine that one is wrong or that one is more right than the other. By using the norming system, assessment creators can improve their craft and learn to write better assessment evaluation materials in the future (Richard, Mark, & Ram, 2007).
Reliability and validity are important in other applications as well. In using the same determining factors, these terms can also be applied to employees when evaluating the level of trust they have earned, the way they treat others and if it is fair, and how engaged the employee is in their work. On the other side, employees can evaluate their supervisors as being reliable if they can be depended upon for fairness and guidance when there are issues to be addressed as well as valid in the ability of the employee to be able to take information from the employer as accurate without needing to verify the given material first-hand.
References
Creswell, J. W. (2005). Educational research: Planning, conducting, and evaluating quantitative
and qualitative research (2nd ed.). Thousand Oaks, CA: Sage Publications. Upper Saddle
River, NJ: Pearson Education.
Pepper, M. B., & Pathak, S. (2008). Classroom contribution: What do students perceive as fair
assessment? Journal of Education for Business, 83(6), 360-367. Retrieved from
http://search.proquest.com/docview/202823952?accountid=35812
Richard, L. P., Mark, L. B., & Ram, B. M. (2007). What is "normal" in faculty norming reports?
The Journal of Applied Business and Economics, 7(4), 1-8. Retrieved from
http://search.proquest.com/docview/218684952?accountid=35812
