Module 3. Psychological Scaling
No matter which popular definition of the term measurement you choose, several underlying themes emerge. First, we need to be able to quantify the attribute of interest. That is, we need to have numbers to designate how much (or little) of an attribute an individual possesses. Second, we must be able to quantify our attribute of interest in a consistent and systematic way (i.e., standardization). That is, we need to make sure that if someone else wants to replicate our measurement process, it is systematic enough that meaningful replication is possible. Finally, we must remember that we are measuring attributes of individuals (or objects), not the individuals per se. This last point is particularly important when performing high-stakes testing or when dealing with sensitive subject matter. For example, if we disqualify a job candidate because he or she scored below the established cutoff on a pre-employment drug test, we want to make sure that the person is not labeled as a drug addict. Our tests are not perfect, and whenever we set a cutoff on a test, we may be making an error by designating someone as above or below the cutoff. In the previous example, we may be mistakenly classifying someone as a drug abuser when, in fact, he or she is not.