A scoring rubric has several components, each of which contributes to its usefulness.  These components include one or more dimensions on which performance is rated, definitions and examples that illustrate the attribute(s) being measured and a rating Scale for each dimension.  Ideally, there should also be examples of student work that fall at each level of the rating scale.

The scoring rubric rating scales may be numerical, qualitative, or a combination of the two.  Qualitative Rubrics may have scale points with labels such as these:

• Not yet, developing, achieving
• Emerging, developing, achieving
• Novice, apprentice, proficient, distinguished
• No evidence, minimal evidence, partial evidence, complete evidence

California's math rubric uses a combination of numerical and qualitative scales:

1. Unable to begin effectively
2. Begins, but fails to complete problem
3. Serious flaws but nearly satisfactory
4. Minor flaws but satisfactory
5. Competent response
6. Exemplary response

Each point on the scale should be clearly labeled and defined. There is no single best number of scale points, although it is best to avoid scales with more than 6-7 points.  With very long scales, it is often difficult to adequately differentiate between adjacent scale points (e.g., on a 100-point scale, it would be hard to explain why you assigned a score of 81 rather than 80 or 82).  It is also harder to get different scorers to agree on ratings when very long scales are used.  The rule of thumb is to have as many scale points as can be well defined and that adequately cover the range from very poor to excellent performance.

How many points should a rating scale have?

There is no one right answer to this question. Consider these as you make your decision:

• Each point on the scale needs to be well defined.  This may be difficult to do for longer scales.
• Longer scales make it harder to get agreement among scorers (inter-rater reliability).
• Extremely short scales make it difficult to identify small differences between students.
• Do you simply want to divide students into two or three groups, based on whether they have attained or exceeded the Standard for an outcome?  If so, then a short scale may be adequate.
• If you are rating a product/performance on several different dimensions, will you want to add up the scores so that each is equally weighted?  If so, you may find it easier to have all scales the same length.