Journal
SEMINARS IN HEMATOLOGY
Volume 45, Issue 3, Pages 150-159Publisher
W B SAUNDERS CO-ELSEVIER INC
DOI: 10.1053/j.seminhematol.2008.04.006
Keywords
-
Categories
Ask authors/readers for more resources
The absolute truth in research is unobtainable, as no evidence or research hypothesis is ever 100% conclusive. Therefore, all data and inferences can in principle be considered as inconclusive. Scientific inference and decision-making need to take into account errors, which are unavoidable in the research enterprise. The errors can occur at the level of conclusions that aim to discern the truthfulness of research hypothesis based on the accuracy of research evidence and hypothesis, and decisions, the goal of which is to enable optimal decision-making under present and specific circumstances. To optimize the chance of both correct conclusions and correct decisions, the synthesis of all major statistical approaches to clinical research is needed. The integration of these approaches (frequentist, Bayesian, and decision-analytic) can be accomplished through formal risk: benefit (R:B) analysis. This chapter illustrates the rational choice of a research hypothesis using R:B analysis based on decision-theoretic expected utility theory framework and the concept of acceptable regret to calculate the threshold probability of the truth above which the benefit of accepting a research hypothesis outweighs its risks.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available