Practical Significance in Special Education Research

Abstract

Whereas statistical significance indicates whether a difference is a difference, additional information signifying whether the difference is large enough to make the results worth acting upon is included in practical significance. Statistical measures of practical significance report the extent to which variation in the dependent variable can be explained by the independent variable. Measures of practical significance are seldom reported in The Journal of Special Education, although sometimes issues of practicality are discussed using the estimate of statistical significance. Practical significance, however, may vary greatly among samples even though statistical significance estimates are similar.

Practical Significance in Early Intervention Research

“…It is not uncommon to find examples of authors stating that a finding is "highly significant" or reporting p values to four or five decimal places, as if something of critical importance were contained in the several zeros preceding the final digit (Feinstein, 1985;Maitland, 1982). Hanson, Abramson, and McNamara (1987) recently demonstrated the fallacy of this assumption in the special education literature. They examined the relationship between magnitude of association measures and probability levels and found that the range of explained variance (10% to 59%) was the same for statistical tests reported significant at the p<.01 and p<.001 levels.…”