Here’s a new article providing a simple example why more than statistical significance is needed for actionable inference #statistics
In this June blog post, we run a through a quick example why statistical significance doesn't tell the whole story and should not be the sole basis for decision making or action. Some main takeaways are: 1. Statistical significance pertains to the existence of a relationship, not the size of it 2. Pay attention to the magnitude of the relationship. This will get you closer to understanding and interpreting whether or not it matters. 3. Translate the magnitudes into real-world implications of using them for decisions or courses of action. Cost-benefit and what-if scenarios can go a long way. #statistics #decisionmaking https://lnkd.in/gy4KHmv7
Biostatistician at 2KMM CRO ⦿ 100% R in Clinical Trials ⦿ Frequentist (non-Bayesian) paradigm ⦿ NOT a Data Scientist (no ML/AI) ⦿ Against anti-{🚗/🥩/💵/🏠} restrictions ⦿ In memory of The Volhynian Mаssасrе OUN-UPA 1943
1moYes, because it's a measure of how something is statistically discernible (detectable), not importance. Luckily we can combine both practical significance and the statistical one through interval hypothesis testing, if only we can specify the minimal meaningful magnitude of the effect that is important to us (not to be confused with measures like Cohen's d effect size etc; except when we deal with standardized outcomes expressed in z-scores). This is exactly how non-inferiority, equivalence and clinical superiority trials work. (here I assume one uses the frequentist framework, but including the information about domain relevance pertains also to the Bayesian approach - inference of neither kind itself accounts to a context.)