Validity is one of the major current issues under consideration, which is specifically relevant to teachers. Essentially, teaching is a profession, which is characterized by factors such as multidimensional, holistic and ever-changing phenomenon. As such, in the attempt to assess teachers’ performance, the process is usually faced with some difficulties (The Center for the Enhancement of Teaching, 2012). Nevertheless, the process of assessment should be such that it is trustable and authentic. This paper is going to explore the issue of validity in assessment and bring out relevant recommendations to improve validity.
Validity in simple terms can be defined as the degree to which a particular assessment tool is able to measure what it is supposed to measure (Cozby, 2009). It is the extent to which an assessment tool measures what it was actually designed to measure, without getting contaminated by other characteristics (Crooks 2010). For instance, if a test is designed for measuring the mathematical ability of the students, then it should only measure mathematical ability and not the ability to read a comprehension. However, most of the time the tool used to assessing education does not always reflect the status of teacher or student performance. In this regard, any tool designed to assess education should have a clear purpose since nothing will be gained unless the process has sufficient validity in itself (Williamson, 2009). It is also important to note that validity has a very close relationship with reliability. Reliability in this case being the ability of a test to produce the consistent results while assessing a particular aspect under varying conditions. In perspective, for an assessment tool to be valid it must first be reliable and provide results that reflect what status of the entire population as opposed to just the selected sample (Musial et.al, 2008). This is the main issue that arises as a result of validity in assessment; whether or not the tool used to assess will produce the results that reflect the entire educational process.
There are different types of validity that are applied in educational assessment. First, there is the content validity, which assures the assessors that the overall sample under study represents the entire content to be studied. This is usually accomplished by accurately selecting test items that accurately represent the content provided in education (Wilkerson & Lang, 2007). However, this is faced by the problem of determining whether the sample selected reflects the entire population or just the sample itself. The second type of validity is criterion validity, which ensures that the test itself is valid and accurate. This can take the form of either concurrent validity where the measure in use correlates with another measure used previously and proved to be valid or predictive validity where the current test can be used to test the validity of a future measure. The other type of validity is the construct validity, which ensures that the assessment tool measures the construct, which it claims to measure (Valencia, & Suzuki, 2010). Generally, validity issues run throughout out the educational process all the way from instruction, curriculum, to assessment. Presence of invalidity in one step affects all the other steps, making the entire education process as well as assessment invalid.
Nonetheless, there a number of ways in which validity in assessment can be improved. First of all, goals and objectives of the assessment process must be clearly defined and operationalized to ensure that all expectations are met (Johnson & Penny, 2008). Secondly, the assessment measure must be matched with the goals and objectives of the process. This will also include involvement of an outside party from other faculty to get feedback from a party that does not directly invest in the instrument (Shepard, 2009). Thirdly, all students should be involved in order to identify an area of great weaknesses and points in the assessment that appear to be unclear. Finally, the measure used as a measurement tool should be compared with other measures and data available to ensure consistency and accuracy of the measurement (Assessment Online, 2013).
In conclusion, it is of vital importance to ensure that the test selected to assess education and factors relating to education have sufficient validity. Failure to do this will translate to incorrect report and consequently inappropriate measure will be taken which can lead to breakdown of the entire educational process.