By: Deborah O'Malley, M.Sc. & Ishan Goel | Last updated February, 2024 Overview:
In this informative 50-minute video interview, Associate Director of Data Science
Ishan Goel of the top testing platform, VWO, answers the key question: can you trust large uplifts in your test results?
Ishan discusses what constitutes a trustworthy test result, based on
statistical significance and sample size requirements.
You'll learn from him:
How confidence intervals impact trustworthy test results. Hint: as more samples are gathered, the confidence intervals become smaller and your data becomes more trustworthy.
The difference between the sample
mean (standard error) and the population mean and why these terms matter to determine sample size requirements. Why you don't necessarily need large sample sizes to get trustworthy test results, as long as the outcome is statistically significant.
Twyman's Law" is, why it's important to know, and how you can apply it to judge whether your own test results are trustworthy. The importance of evaluating your test results based on past history, data segmentation, and consistency.
Ishan offers a wealth of knowledge and actionable insights.
Listen and learn from one of the foremost data science experts in the field.