In this article, the authors discuss how the misuse of norm-referenced tests can impact the assessment and treatment of a client. Norm-referenced tests provide a comparison between the skills and behaviors assessed of a client to the relevant norms of a similar age group. According to the article, a clinician must ensure to properly use a norm-referenced test in order to provide evidence as to whether a client may need more assessments or whether a certain treatment approach is more beneficial to the client. However, the misuse of a norm-referenced test may also negatively impact the client’s diagnosis and treatment approach. In this article, the authors describe four common errors that arise when misusing a norm-referenced test.
As previously mentioned, norm-referenced tests are used as a comparison in order to evaluate the client’s scores. However, the authors discuss that norm referenced tests are not truly reliable as comparisons are based on estimates and no two similar individuals will perform in an identical matter on a given task. This leads to the question of how much of a difference should there be between a client and similar individuals in their age group before the differences are indicative of an underlying problem? The use of reliability may assist clinicians in determining this issue. The authors describe that depending on how high or low the tests reliability is, is an indication of whether there is little difference or inconsistencies between the client’s ideal score, also known as their true score, and their observed score.
Due to the issue that norm-referenced tests are based on estimates, a test’s reliability may cause doubt for a clinician to interpret a client’s test score results. Therefore, the estimated true score and a confidence interval may be used to estimate the accuracy of what the tests was designed to measure. The confidence interval is based on the standard error of measurement (SEM). In order to find the SEM, which will allow one to find the confidence interval, the test reliability estimate, normative sample’s mean and standard deviation scores, and the client’s observed score is needed.
Another issue that may arise when using a norm-referenced test is that a client may have a variation in their scores that may or may not be indicative of a problem. Standard deviations and z scores are used in order to determine if a test...