Statistically Speaking Ep 4: Level of Significance & P-values (Sounds Boring But Stay With Us)

saraboostani

I've been posting a bit about our video series called Statistically Speaking — a labor of love from the research team at 6sense.

Ever since day one of this job, we've been committed to producing objective, thorough research. It's kind of in our blood — or at least how we were taught. That means crossing t's and dotting i's with regressions, ANOVAs, and all the other tests you may vaguely remember from your own college stats classes.

Far too often, we see reports confidently declaring that Thing 1 is the winner and Thing 2 is the loser with no statistical backing and a 2 percentage point difference between them — however, that difference could simply be random chance. And yet real business decisions get made from them. X was just shown in [Report] — we should be doing that. Deliberate deception? I highly doubt it. But it matters — whether you're someone who produces research or simply consumes it (all of us).

It turns out what your stats professor was trying to teach you in that under-airconditioned lecture hall was actually relevant. In this series, we refresh your memory on the statistical concepts you need when reading research — and help you spot when they're being used well or if they're being used at all.

In this episode: level of significance and p-value. Sound familiar? Walk away with a clear understanding in less than 20 minutes. Real research examples and practical application.

Watch the full episode here!

https://6senseinc.wistia.com/medias/cnuh7gvd7y?embedType=web_component&seo=true&videoWidth=960

@Kerry