← EDU 311 Educational Statistics
Measures of Dispersion in Classroom Testing
Notes
Understanding Score Dispersion
Measures of central tendency indicate the general performance of a class, but they do not reveal how scores vary among learners. Two classes can achieve the identical mean score of sixty percent, yet one class may consist of evenly performing students while the other contains both top achievers and failing students. In educational measurement, dispersion describes the extent of spread, scatter, or variability among assessment scores.
Key Concepts
Range: The simplest measure of variability, computed by subtracting the lowest score from the highest score. While quick to compute, the range relies exclusively on two extreme values and ignores all intermediate scores.
Variance: The mean of the squared deviations of scores from their arithmetic mean. Variance quantifies how far individual learners deviate from the group average. Because deviations are squared to eliminate negative signs, variance is expressed in squared units, which can be difficult to interpret directly against test marks.
Standard Deviation: The positive square root of the variance. Standard deviation is the most reliable and widely reported indicator of test variability in educational research and Kenya National Examinations Council analyses. It reflects the average distance that scores deviate from the mean in original test units.
Significance of Standard Deviation
A small standard deviation signifies a homogeneous class where students perform at comparable ability levels. In such classrooms, whole-group instruction is effective. A large standard deviation indicates a heterogeneous class with widespread ability differences. This requires the teacher to implement differentiated teaching, group work, and individualized remedial activities to cater to diverse learning paces.
Practical example
Madam Akinyi teaches two Form Two streams at Kisumu Girls High School. Both streams sat for a common Biology assessment marked out of 20. She wants to determine which class is more heterogeneous using a sample of five students from each stream.
Stream A scores: 10, 12, 14, 16, 18. Mean equals 14.
Deviations from mean: minus 4, minus 2, 0, 2, 4.
Squared deviations: 16, 4, 0, 4, 16. Sum of squared deviations equals 40.
Sample variance equals 40 divided by 4, which is 10.
Standard deviation equals the square root of 10, which is approximately 3.16 marks.
Stream B scores: 6, 10, 14, 18, 22. Mean equals 14.
Deviations from mean: minus 8, minus 4, 0, 4, 8.
Squared deviations: 64, 16, 0, 16, 64. Sum of squared deviations equals 160.
Sample variance equals 160 divided by 4, which is 40.
Standard deviation equals the square root of 40, which is approximately 6.32 marks.
Madam Akinyi observes that while both streams share an identical mean of 14, Stream B has double the standard deviation of Stream A. Stream B is far more heterogeneous and requires differentiated instruction.
Quiz
1. What does a very low standard deviation in a KCSE trial examination indicate about a school cohort?
2. How is the sample standard deviation mathematically derived from the variance?
3. A teacher finds that five students scored 40, 40, 40, 40, and 40 on a diagnostic quiz. What is the standard deviation of this data set?
Sign in to submit and securely grade your quiz.