|
|
1.
|
|
What it there were no significance tests? / Lisa Harlow, co-aut. Stanley Mulaik, co-aut. James Steiger Nível de parte analítica: Problem is epistemology, not statistics : replace significance tests by confidence intervals and quantify accuracy of ... • Good science is abductive, not hypothetico-deductive • Testing small, not null, hypotheses : classical and bayesioan approaches • Introduction to bayesian inference and its applications • When confidence intervals should be used instead of statistical tests, and vice versa • Noncentraly interval estimation and the evaluation of statistical models • Goodness of approximation in the linear model • Case study in the failure of psychology as a cumulative science : the spontaneous recovery of verbal learning • Reforming significance testing via three-valued logic • Retrospective on the significance test ban of 1999 : if there were no significance tests, they would be invented • There is a time and a place for significance testing • Eight common but false objections to the discontonuation of significance testing in the analysis of research data • Earth is round -p<.05- • Significance testing and overviewPublicação: Mahwah : Lawrence Erlbaum, 1997Descrição: 446 p.Disponibilidade:
|
