Statistics That Survive the Meeting: Practical Statistical Analysis for Better Questions, Honest Uncertainty, Stronger Models, and Sound Decisions

Statistics That Survive the Meeting Practical Statistical Analysis for Better Questions, Honest Uncertainty, Stronger Models, and Sound Decisions

A chart can be correct and still lead to a bad decision.

A p-value can be technically valid and still answer the wrong question. A model can predict well for months, then fail when the population changes. An average can improve while the people at the edge of the distribution have a much worse experience.

Statistics That Survive the Meeting is a practical guide to the judgment that sits between raw data and a defensible decision. It is written for analysts, managers, researchers, operators, consultants, and anyone who has ever looked at a dashboard and wondered, “What can this number actually support?”

Instead of burying the reader in formulas, Camille Panganiban focuses on the parts of statistical work that most often decide whether an analysis holds up under scrutiny: framing the question, choosing the right comparison, understanding the unit of analysis, reading uncertainty, checking assumptions, and connecting results to real decisions.

Inside, you will learn how to: - Turn vague questions such as “Is this working?” into testable statistical questions. - Separate a descriptive change from evidence that an intervention caused it. - Choose comparison groups that match the claim you want to make. - Read confidence intervals and other uncertainty measures without treating them as decorative footnotes. - Recognize when averages hide meaningful variation or subgroup effects. - Spot nonlinear relationships before a straight-line model tells a crooked story. - Use interactions and multivariable models without stuffing every available column into the analysis. - Distinguish prediction from explanation and forecasting from causal inference. - Build one-page question, evidence, and result templates that make analytical judgment visible. - Design statistical practices that survive deadlines, handoffs, and inconvenient results. - Catch recurring failure modes such as confirmation bias, model drift, silent assumptions, and misleading aggregation. - Translate a statistical result into a decision rule, a follow-up measure, and a clear condition for changing course.

The book uses realistic business, healthcare, logistics, product, pricing, and operational examples to show what good statistical reasoning looks like when the data is messy and the meeting is already on the calendar.

There is also a healthy amount of skepticism. Software will happily calculate a p-value for nonsense. Dashboards can put a green arrow beside a fragile result. A polished slide can make a weak comparison look oddly confident. The answer is not cynicism. It is a better analytical process.

By the end, you will have a repeatable way to move from question to evidence, from evidence to interpretation, and from interpretation to action without asking the numbers to promise more than they can deliver.

If you want statistical analysis that remains useful after the spreadsheet closes and the questions start, Statistics That Survive the Meeting belongs on your desk.

© 2026 Mission 333 Press (E-bog): 6610001378714

Udgivelsesdato

E-bog: 17. september 2026