Category Archives: Statistics

The Importance of Statistical Power in Online A/B Testing

Statistical Power and Test Sensitivity

What is Statistical Power? In null-hypothesis statistical testing (NHST) – the procedure most commonly applied in A/B tests, there are two types of errors that practitioners should care about, type I and type II errors. Type I is the probability of the test procedure to falsely reject a true null hypothesis. Type II error is […] Read More…

Also posted in A/B Testing, AGILE A/B Testing, Statistical Significance | Tagged , , , , , | Leave a comment

5 Reasons to Go Bayesian in AB Testing – Debunked

Frequentist vs Bayesian A/B Testing

As someone who spent a good deal of time on trying to figure out how to run A/B tests properly and efficiently, I was intrigued to find a slide from a presentation by VWO®’s data scientist Chris Stucchio, where he goes over the main reasons that caused him and the VWO® team to consider and […] Read More…

Also posted in A/B Testing, Conversion Optimization, Statistical Significance | Tagged , , , , , , | Leave a comment

The Bane of AB Testing: Reaching Statistical Significance

Illusory Results AB Testing

Now, this is not a new topic on this blog. I’ve discussed the issue of optional stopping based on “achieving” or “reaching” statistical significance in my Why Every Internet Marketer Should be a Statistician post more than two years ago and others have touched on it as well – both before and after me. However, the issue […] Read More…

Also posted in A/B Testing, Conversion Optimization, Statistical Significance | Tagged , , , , , , , , , , | Leave a comment

Bayesian AB Testing is Not Immune to Optional Stopping Issues

Fantasy vs real world in bayesian ab testing

Fantasy vs the Real World: Naive Bayesian AB Testing vs Proper Statistical Inference This post is addressed at a certain camp of proponents and practitioners of A/B testing based on Bayesian statistical methods who claim that outcome-based optional stopping, often called data peeking or data-driven stopping, has no effect on the statistics and thus inferences […] Read More…

Also posted in A/B Testing, Conversion Optimization, Statistical Significance | Tagged , , , , , , , | Leave a comment

Overview of Books on Conversion Optimization, Web Analytics, SEO & PPC

This overview is a little different than most overeviews out there. It will focus on three specific basic statistical concepts and would show if and how well they are explored in 18 of the top books on Conversion Optimization, A/B testing, web analytics, SEO & PPC/AdWords. The three concepts this review focuses on are statistical […] Read More…

Also posted in A/B Testing, Conversion Optimization, Google Analytics, Internet Marketing, Multiple Variations Testing, Statistical Significance | Tagged , , | Leave a comment

Should you do A/A, A/A/B or A/A/B/B tests in CRO?

A/A Split Test

This question pops up often in more “advanced” discussion forums and blogs on Conversion Optimization and usually one (or several) of the following advices are given: Do an A/A test first in order to test your split testing framework. If the difference between the two is statistically significant at the decided level, then your framework is […] Read More…

Also posted in A/B Testing, Conversion Optimization | Tagged , , | Leave a comment