The Measurement of Scores By Criterion and Non-Criterion Interpretation
Measurement is the process of assigning numbers to represent student achievement or performance according to certain rules, for instance, answering 85 out of 100 items correctly on a test. The numbers or scores indicate the degree to which a learner possesses a certain characteristic or trait (Brookhart & Nitko , 2008).
Measurement is important for reporting the achievement of learners on nursing and other tests, but not all outcomes important in nursing practice can be measured by testing. Many outcomes are evaluated qualitatively through other means, such as observations of performance.
Although measurement involves assigning numbers to reflect learning, these numbers in and of themselves have no meaning. Scoring 15 on a test means nothing unless it is referenced or compared with other students’ scores or to a predetermined standard.
Perhaps 15 were the highest or lowest score on the test, compared with other students. Or the student might have set a personal goal of achieving 15 on the test; Thus meeting this goal is more important than how others scored on the test.
Another interpretation is that a score of 15 might be the standard expected of this particular group of learners. To interpret the score and give it meaning, having a reference point with which to compare a particular test score is essential. In clinical practice, how does a learner’s performance compare with that of others in the group?
Did the learner meet the clinical objectives and develop the essential competencies regardless of how other students in the group performed in clinical practice? Answers to these questions depend on the basis used for interpreting clinical performance, similar to interpreting test scores.
Norm-Referenced Interpretation of Test Score
There are two main ways of interpreting test scores and other types of assessment results: norm-referencing and criterion-referencing. In norm-referenced interpretation, test scores and other assessment data are compared to those of a norm group. Norm-referenced interpretation compares a student’s test scores with those of others in the class or with some other relevant group.
The student’s score may be described as below or above average or at a certain rank in the class. Problems with norm-referenced interpretations, for example, “ grading on a curve ,” are that they do not indicate what the student can and cannot do, and the interpretation of a student’s performance can vary widely depending on the particular comparison group selected.
In clinical settings, norm-referenced interpretations compare the student’s clinical performance with those of a group of learners, indicating that the student has more or less clinical competence than others in the group. A clinical evaluation instrument in which student performance is rated on a scale of below to above average reflects a norm referenced system. Again, norm-referenced clinical performance does not indicate whether a student has developed desired competencies, only whether a student performed better or worse than other students.
Criterion Referenced Interpretation
Criterion-referenced interpretation, on the other hand, involves interpreting scores based on preset criteria, not in relation to the group of learners. With this type of measurement, an individual score is compared to a preset standard or criterion. The concern is how well the student performed and what the student can do regardless of the performance of other learners.
Criterion-referenced interpretations may:
(a) describes the specific learning tasks a student can perform, for example, define medical terms
(b) indicate the percentage of tasks performed or items answered correctly, for example, define correctly 80% of the terms
(c) compare performance against a set standard and decide whether the student meets that standard, for example, met the medical terminology competence (Miller, Linn, & Gronlund, 2009).
Criterion-referenced interpretation determines how well the student performed at the end of the instruction in comparison with the objectives and competencies to be achieved. With criterion- referenced clinical evaluation, student performance is compared against preset criteria.
In some nursing courses these criteria are the clinical objectives to be met in the course. Other courses indicate competencies to be demonstrated in clinical practice, which are then used as the standards for evaluation.
Rather than comparing the performance of the student to others in the group, and indicating that the student was above or below the average of the group, in criteria referenced clinical evaluation, performance is measured against the objectives or competencies to be demonstrated.
The concern with criterion-referenced clinical evaluation is whether students achieved the clinical objectives or demonstrated the competencies, not how well they performed in comparison to the other students.