@@lovehoney9949 this is one of those questions where I think the best answer is that it depends. It depends on the sample size you have, the effect size and really the implications of the business/practical decision you are making. I think there are some cases where I would say yes, we should consider practical significance even if we don’t have statistical significance. Especially when intuitively we are convince the decision makes sense. It then becomes an art and science. However if the consequences of the decision are risky like concluding someone is sick, needs an invasive treatment, millions of dollars will be lost based on the decision, etc. it would be worth confirming statistical significance even first or knowing at what confidence level the conclusion would be statistically significance at .
Should we consider practical significance in cases where we don’t have statistical significance?
@@lovehoney9949 this is one of those questions where I think the best answer is that it depends. It depends on the sample size you have, the effect size and really the implications of the business/practical decision you are making. I think there are some cases where I would say yes, we should consider practical significance even if we don’t have statistical significance. Especially when intuitively we are convince the decision makes sense. It then becomes an art and science. However if the consequences of the decision are risky like concluding someone is sick, needs an invasive treatment, millions of dollars will be lost based on the decision, etc. it would be worth confirming statistical significance even first or knowing at what confidence level the conclusion would be statistically significance at .