First Things First: Verifying Assumptions Before Hypothesis TestingJune 24, 2025 •3:00 PM UTC
When you run a hypothesis test, you’re trusting that your data meets some critical assumptions — but what if it doesn’t? Many practitioners unknowingly apply statistical tests to time-ordered data that violates the assumption of independence, leading to misleading conclusions and costly mistakes.
For example, two processes might both experience trends over time — one drifting upward, the other stationary — yet a t-test could misleadingly suggest no difference. Without first checking for stability and stationarity, you risk trusting results that don’t reflect reality.
Join us for this educational webinar as we explore the critical first step before running any hypothesis test: verifying assumptions. Through real-world examples, we’ll demonstrate how simple graphical checks can reveal hidden patterns like trends, shifts, and instability — and why checking for independence and identical distribution is essential for meaningful analysis.
What You'll Learn
- Understand the key assumptions behind standard statistical analyses and graphs — and why they matter.
- Learn simple tools you can use to test whether critical assumptions like independence and stationarity are violated.
- How to visually detect trends, shifts, and non-random patterns that could compromise your results.
Master Black Belt • MoreSteam Client Services
Kevin Keller is a professional statistician and quality professional with over 30 years of experience teaching, coaching, and leading Lean Six Sigma project teams from the shop floor to the enterprise levels. He began his career as a process engineer at Texas Instruments and then served in a Master Black Belt role at MEMC Electronic Materials for 15 years. Kevin advanced to manage quality systems for AB‐InBev and eventually accepted a Master Black Belt position for AB‐InBev's North American Zone. Kevin earned a BS in Chemical Engineering from Missouri University of Science and Technology and a Masters in Applied Statistics from The Ohio State University. He also is a certified Lean Six Sigma MBB.
