Maine Stanines

Taken from R.E.A. Report – August 10, 2001

Maine Stanines are norming information for first grade students in Maine collected in the 1995-1996 academic year for the Observation Survey. Stanines allow for comparisons to be made across different tests and provide a picture of where the students in the sample fall relative to the norming group (in this case, Maine first graders). Kindergarten data collected in the spring of kindergarten compares with the first grade fall stanines. First grade data collected in the spring of first grade compares with the first grade spring stanines.

Stanines range from 1 to 9, with 9 being the highest possible score. Students performing in stanines 1 to 3 are performing below average relative to the norm group; stanines 4 to 6 represent average performance (54% of students will fall in this range); stanines 7 to 9 reflect above average performance. Even though stanine 4 represents an average performance relative to the norm group, the criterion for Maine students at year-end is stanine 5. Therefore, students are categorized in the stanine-related tables in one of four performance categories relative to the criterion stanine 5: Below (Stanine 1 to 3), below but On grade level (Stanine 4), at the Criterion (Stanine 5), or Above (Stanine 6 to 9). Note that stanines are a norm-referenced index, therefore it is only possible to use them to compare students with one another; they are not useful to measure students’ actual mastery of the subject matter.

Floor and Ceiling Effects

Maine stanines are norming information collected on first grade students in Maine Reading Recovery schools in fall and at year-end. As standard scores, there is an implication that the distribution of scores is normally distributed with the shape of a bell curve. This is not the case, however, for many of the Observation Survey sub-tests.

Due to a floor effect, some test distributions are positively skewed. This means that most of the students score at the lowest end of the possible range of test scores, while a minority achieves higher scores. In this case, there is not much discrimination among low-scoring students. For example, on the Word Test taken in the fall of first grade (or spring of kindergarten), most students score between 0 and 7 (stanine 4 to 6) with the typical student scoring 3 or 4 (stanine 5). Since 0 is the lowest score, this places a student who scores 0 at stanine 4 rather than at stanine 1.

On the contrary, due to a ceiling effect, some test distributions are negatively skewed. This means that most of the students score at the highest end of possible range of test scores, while a minority achieves lower scores. In this case, the test does not allow high achieving students to show their potential ability. In other words, there is not much discrimination among high-end scorers. For example, on the Word Test taken in the spring of first grade, most students score between 16 and 20 (stanine 4 to 6) with the typical student scoring 17 or 18 (stanine 5). Since 20 is the highest score, this places a student at stanine 6 rather than at stanine 9.

In summary, due to the distribution of a test score being skewed left (negatively) or right (positively), there is not enough variance in scores to cover all 9 stanines. In this case, you will notice that some Observation Survey sub-tests do not contain test scores across the full range of stanines.