Fisher least significant difference test

a statistical procedure to compare pairs of means, conducted after an F test has revealed that at least one pair of means is significantly different. The test calculates the smallest value that would be significantly different from chance when subtracting one mean from another mean. If the absolute value of the actual difference between a pair of means is larger than this least significant difference (LSD), a researcher can reject a null hypothesis that the means are equal and conclude that they are significantly different. Also called protected t test. [Ronald Aylmer Fisher]

Browse dictionary by letter

a b c d e f g h i j k l m n o p q r s t u v w x y z

Psychology term of the day

January 23rd 2025

verbal report

see reportability.