Paper #1480

Títol:

On the failures of the null-hypothesis test

Autor:

Nicholas Longford

Data:

Abril 2015

Resum:

This report comprises four reactions to the recent policy statement in Basic and Applied Social Psychology that announced a ban on null-hypothesis testing in that journal. A personal perspective is presented which agrees with the editors that null-hypothesis testing has become dysfunctional, but proposes a solution different from the editors'. In particular, application of formal statistical methods is defended, but they have to be tailored much more closely to the specifics of the problem that is addressed.

Paraules clau:

decision theory; expected loss; hypothesis testing; plausible value; research agenda.

Codis JEL:

MSC classification: 62F-03, 62G-10, 62H-15.

Descarregar el paper en format PDF