Have you seen a statistical claim and wonder where it’s coming from? Or do you naturally take it at face value without questioning it?
This is why I want to talk about statistical inference.
🗣 Statistical inference is about the sampling of data from a population (or in other words “the world we live in”) AND THEN making “informed guesses” about the world (in extrapolation of the results from the sample of the population).
How intuitive is statistical inference?
What to account for in statistical inference?
👉 Recognize there is sampling variation and presentation of results should be done within context.
👉 The confidence interval should in some way capture a level of true percentage range that the sample represents for the population (ie. the “plausible values” of the population)
👉 Sample size matter, yes. However, is more data better?
Challenge the status quo (ie. often designated as the Null Hypothesis). Can you “throw it out” in favor of new knowledge (ie. the Alternate Hypothesis)?
… there’s so much more to talk in terms of this …leaving more on this subject for another post, because hypothesis testing is the ultimate “hallmark” of experimentation.
👉 With any probability mathematical equation, it’s important to understand the assumptions under which that problem is being solved!
👉 Be aware of a “significance level” reasonable enough to accept in order to reject the null hypothesis. Meaning, what’s the level of randomness willing to be accepted in a world of constant variation, for a particular scenario.
This “artificial threshold” can change according to context and can be very subjective.
Check out what others may have said about this subject on Linkedin