Banded Scoring
With banded scoring, when you receive notification of how you performed on the exam, you will not be given a numerical score (i.e., 67 out of 80, 93 out of 100). Rather, you will be informed into which band (i.e., 1, 2, or 3) your score fell. The following information is provided to help you understand the banding procedure.
What is banding?
Banding is one way to reduce the impact of fluctuations in test scores that do not provide meaningful information about differences in the ability to perform the job. One important purpose of testing is to identify the differences in test scores that reflect real differences among candidates. Banded scoring is a statistical procedure for grouping raw test scores that statistically are not meaningfully different from one another. In banded scoring, bands are set objectively and statistically. They are not manipulated arbitrarily.
Misconception about banding.
There are many misconceptions about banding and the use of banded scores. Some of the most common misconceptions are listed below. Each misconception is followed by a clarification.
Misconception: Each band should have the same number of people.
We do not force bands to be a certain size. The people in a band are similar to each other in that statistically there is no meaningful difference in their scores. Sometimes Band 1 may be very large, and at other times it may be small. People’s scores determine the size of the bands. We never know how many people will be in each band until we receive the test scores.
Misconception: Band numbers have no meaning. I don’t have a score.
Band numbers do have meaning. Think of a band as a group of tied scores. Consider that in school two students with average grades of 94.5 and 94.3 would both be grouped into the same band. Just because one student made a 94.5 and one student made a 94.3, the teacher cannot be sure that 0.2 of a point means that the student who scored 94.5 is smarter or is a better student. The scores are so close to each other that they are basically the same.
For example, think of the achievement tests that children take in school. The fine print on these tests always informs you not to focus on the numerical score but rather on the comparative score, which uses some type of grouping technique such as percentiles, standard deviations, grade levels, etc. These grouping techniques are considered forms of banding. Banding compares your performance on the test to the other test takers’ performance and groups your score with others that are statistically the same.
Misconception: Band numbers are the same as letter grades.
Band numbers are not the same as letter grades. Band 1 does not equate to an “A,” Band 2 to a “B,” and so on. In school, a predetermined numerical range of scores (i.e., 90-100, 80-90, 70-80) equals an alphabetical value (i.e., A, B, C). This grading system is a form of banding. In this case, unlike grade school, the width of bands is not set in advance. Scores are banded only in relation to one another, so you compete against other test takers. The scores of all test takers determine the width of the bands, and your score is set in relation to the scores of your peers.
Misconception: A banded score on one test has the same value as a banded score on another test.
Banded scores are test specific and cannot be compared from test to test. Consider that a test taker scored 88 on one test, and the highest score of all test takers was 89. It is likely for this exam that the test taker who scored 88 would be in Band 1. However, if the same person scored the same grade on another test, and the highest score of all test takers was 100, he/she may be in Band 2 or Band 3. Candidates’ scores vary on each test, and since candidate scores determine the width of bands and into which band test takers fall, the value of a banded score varies from test to test.
Misconception: People with the most seniority who have been on the job longest should be in the top bands.
People with the most experience do not always fall into the top bands. Time spent in a job may not be the same as possessing a knowledge, skill, or ability needed to perform the job. The people with the strongest knowledge, skills, and abilities (or who did best on the exam) will be in the top bands. Some of the people in the top bands will have been in similar jobs for a long period of time, and others will have been in similar jobs for a short period of time. Years of service do not always equal proficiency. Candidates with seniority or experience do not automatically perform best on the test. Regardless of seniority, candidates who display the appropriate knowledge, skills, and abilities perform best on the test.
Misconception: A standing in Band 4 or below automatically indicates failure or ineligibility for jobs.
A band number of 4 or lower is not automatically equated with failure. For one test, there may only be 4 bands, and for another test, there may be 14 bands. So, your success on the test based on your position in a band varies from test to test. Your standing in a band does not indicate whether or not you pass or fail the test. The true test of success in your employment opportunities is whether or not you can be certified and considered for a job vacancy.
Misconception: Banding replaced the “Rule of 10.”
Banding did not replace the “Rule of 10.” The “Rule of 10” determines the number of bands to be certified. In the past, tied scores referred to an actual numerical score (e.g., two candidates with a score of 98.98 were considered tied) while now all of the scores within a band are considered tied.
Misconception: People in a band do not differ.
When several people are placed in the same band, it does not mean that those people do not differ at all. Instead, it means that their scores on the exam do not differ enough to be separate scores.