Cutoff Scores

Uses of Cutoff Scores

One use of cutoff scores may be used to define performance levels by researchers. Generally they are measured numerically and then placed on the corresponding level. Examples of these may be seen in course syllabi grading criteria. The first component of such standards is the letter grades (e.g., A, B, C), which could be further segmented (e.g., A, −A, +B, B). The second, parallel component would be their numerical equivalents (e.g., A+ = 100%, A = 99–93%, A− = 92–90%). The role of the first and second segments is to predict the level of competence an individual has in a specific course. Using this information, one might guess that an individual who earns scores on the higher end of the grading spectrum might perform well in the workforce. Likewise, individuals who earn lower scores might not perform as well in the workforce. Such academic standards and their corresponding cutoff scores have been used for a long time. There may be several implications of this evaluation system. On one hand, this system generally works; individuals earn good marks and once they enter the workforce, they typically need minimal training. Such findings may be useful to potential employers who have a need for employees but lack the financial resources to further train those who did not earn good grades and enter the workforce as competent as their higher-scoring peers.

The preceding examples are quite simplistic, but since this entry is providing a basic explanation of cutoff scores, college undergraduate students and their experiences will be used to demonstrate the implications of cutoff scores. The examples to follow will involve study abroad experiences and the issue of cultural competency, which is the level of sensitivity one has toward another’s culture. For example, an American might offend an individual from Saudi Arabia if he were to extend his left hand (as opposed to his right) to shake hands. This would demonstrate a low cultural competence.

Scholars may use a norm-referenced framework to understand a specific population. As an example, let us examine students and perceived competency. However, let us assume these students are undergraduates who are participating in a faculty-led study abroad program. A scholar might want to test if the students become culturally competent at the conclusion of the trip. Generally for a norm-referenced framework, researchers will need to assess a large population. In this example, it would be ideal if all students who studied abroad filled out a survey or questionnaire concerning their experience. The researcher may attempt to assess all the students who participated in a study abroad program at his or her university during one academic year. Should the entire student population who studied abroad create a large enough sample size to determine the effect of the trips on cultural competency, then it can be concluded that the result is the norm for that population. Normed framed references, unlike criterion-framed references, are largely straightforward if a large enough sample size can be obtained.

...