It’s our first stats reunion! In this special review episode, we revisit favorite concepts from past episodes—p-values, multiple testing, regression adjustment—and give them fresh personalities as characters. Meet the seductive false positive, the clingy post hoc ex, and Charlotte, the well-meaning but overfitting idealist.

Statistical topics

Bar charts vs Box plots
Bonferroni correction
Confounding
False positives
Multiple testing
Multivariable regression
Outcome switching
Over-adjustment
Post hoc analysis
Pre-registration
Residual confounding
Statistical adjustment using regression
Subgroup analysis
Unmeasured confounding

Review Sheet

References

Nuzzo RL. The Box Plots Alternative for Visualizing Quantitative Data. PM R. 2016 Mar;8(3):268-72. doi: 10.1016/j.pmrj.2016.02.001. Epub 2016 Feb 15. PMID: 26892802.
Sainani KL. The problem of multiple testing. PM R. 2009 Dec;1(12):1098-103. doi: 10.1016/j.pmrj.2009.10.004. PMID: 20006317.

Kristin and Regina’s online courses:

Demystifying Data: A Modern Approach to Statistical Understanding

Clinical Trials: Design, Strategy, and Analysis

Medical Statistics Certificate Program

Writing in the Sciences

Epidemiology and Clinical Research Graduate Certificate Program

Programs that we teach in:

Epidemiology and Clinical Research Graduate Certificate Program

Find us on:

Kristin - LinkedIn & Twitter/X

Regina - LinkedIn & ReginaNuzzo.com

(00:00) - Intro
(02:26) - Mailbag
(06:42) - P-values
(12:43) - Multiple Testing Guy
(16:05) - Bonferroni solution
(17:11) - Post hoc analysis ex
(22:22) - Subgroup analysis person
(30:28) - Statistical adjustment idealist
(43:54) - Unmeasured confounding
(45:19) - Residual confounding
(49:25) - Over-adjustment
(54:42) - Wrap-up

Stats Reunion: What have we learned so far?