Discussion
The word "noir" is used to remember the scaling of measurement in psychology (Embretson, 2004). In short, the letters stand for nominal, ordinal, interval and ratio (Embretson, 2004). To give a brief introduction of what each scale measures, "nominal is the simplest way to measure" because it focuses on categorizing measurements on a scale of category, according to Embretson (2004). An example of nominal is eye color.
"Ordinal measures in terms of ranking, interval measures scores of tests that focus on unobservable mental functioning and ratio focuses on measuring activities in the physical world, such as someone's running time" (Embretson, 2004). With different scales of measurement, there are two methods to compare sets of data.
These include norm-referenced and criterion-referenced testing. According to Embretson (2004) norm-referenced testing "yields information on a testtaker's standing or ranking relative to some comparison group of testtakers." In other words, it focuses on the performance of peers. Criterion-referenced testing is a little different because it focuses on examining individual's scores to a set standard (Embretson, 2004).
The ability for ordinal measurement scale to be utilized on a standardized test as a norm-referenced test is high since an ordinal scale is based upon ranking and norm-referenced testing gathers information on the examinees ranking compared to a group of testtakers. For example, a study conducted on decision making with the use of ordinal variables states that ordinal measurement scales has the ability to be utilized by norm-referenced testing (Barua, Kademane, Das, Gubbiyappa, Verma, & Al-Dubai, 2014).On the other hand, ordinal scaling would not be a strong measurement for criterion-referenced testing because it focuses on the ranking rather than the measurement of the scores to be close to a set standard.
Ratio scaling directs its focus on measuring objects and activities in the physical world which would be beneficial for criterion-referenced testing instead of norm-referenced testing. Imagine a marathon runner who was trying to beat the world's fastest time running a marathon. Criterion-referenced testing allows the runner to be aware of the set standard the marathon runner needs to beat to be the best and set a new standard.
Norm-referenced testing would not be as useful because the marathon runner would not have the standard measurement he or she needs to beat. However, the marathon runner would be aware of the relative time he or she needs to beat to be the best. That is not as helpful as the criterion-referenced testing because runners need an exact number instead of a relative number in comparison to other runners.
Norm-referenced data would be collected by "the standards relative to a group, such as means and standard deviations" while criterion-referenced data is collected "based on absolute standards" (Montgomery & Connolly, 1987). Using "noir" as the scale of measurement in psychology is significant because nominal and ordinal are thought of as categories and interval and ratio are ways to collect quantitative data. Therefore, "choosing a statistical analysis procedure depends on the level of measurement of variables in the experiment" (Virginia Tech, 1999).
In other words, the flexibility and rigidity of this particular scale of measurement in psychology allows professionals have the ability to measure in terms of categories or quantitatively. For my career as a BCBA, I believe the norm-referenced testing is the most appropriate use of measurement for standardized testing because I am looking for my clients to demonstrate a relative understanding on speech, motor skills, social skills and leisure skills.
Other tests in school may require my clients to reach a set standard of testing such as end of the year testing to move on to the next grade or submitting scores for the SAT to get into college, but my goal as a BCBA is to help clients generalize the information relatively compared to their peers on a normal functioning level.
References
Barua, A., Kademane, K., Das, B., Gubbiyappa, K. S., Verma, R. K., & Al-Dubai, S. (2014).
A tool for decision-making in norm-referenced survey questionnaires with items of ordinal variables. International Journal of Collaborative Research on Internal Medicine & Public Health, 6(3), 52-63.
Embretson, S. E. (2004). The Second Century of Ability Testing: Some Predictions and Speculations. Measurement, 2(1), 170-174.
Montgomery C. P. & Connolly H. B. (1987). Norm-Referenced and Criterion-Referenced Tests: Use in Pediatrics and Appliation to Task Analysis of Motor Skill. ResearchGate 67(12), 1873-1875.
Virginia Tech. (1999). Measurement Scales in Social Science Research.