When creating or interpreting a test, it is important to consider to whom you will compare your examinees. What is the appropriate population for creating norms? For example, as people age, they tend to get lower scores on tasks requiring speed. Does it make more sense to use age-related norms or should the examinees be assigned a percentile that reflects their performance relative to the general population? There is no one-size-fits-all answer. If the test is to be used to detect the possibility of brain disease, then one would want to score relative to age norms to prevent false positives in older test-takers. If the test is to be used to evaluate whether one can safely operate a machine requiring quick reaction times, then one would use general norms.
We are focused on norms and scaling test scores for interpretation. Test scores can be normed in a number of ways, and the approach taken has implications for interpretation. Those familiar with things like the MMPI may be familiar with the term “spike” on its scales. The “spike” is based on the responses to items compared to “normal people”. This comparison is what makes the MMPI useful for clinicians. In addition, this week you will be doing some basic calculations in the course dataset in the context of investigating norms, and the use of norms in interpreting data.
To prepare for this Discussion, consider the use of norms when making decisions about individuals in organizations.
With these thoughts in mind:
Post by Day 4
Explain the use of norms in
Explain the pros and cons of group level norms in test interpretation