Marathon Performance: Does high-carb fueling work?

February 23
1h 5m

Episode Description

How many carbs do you need to run your best marathon? Recent headlines suggest that 120 grams per hour is the magic number. But what’s the science behind that claim? To find out, we dug into the study fueling the hype — and were surprised by what we found. In this episode, we uncover numbers that mysteriously shift after peer review, figures that don’t match the text, and p-values that refuse to line up with their confidence intervals. Along the way, we swap bonking stories, revisit repeated-measures ANOVA, renew our antipathy for spreadsheets, and follow a trail of statistical termites to a surprisingly happy scientific ending.


Statistical topics

  • Article in press vs final publication
  • Data management and workflow
  • Multiple testing
  • P-values and confidence intervals
  • Repeated Measures ANOVA
  • Statistical sleuthing
  • Version control in research
  • Within-person study design


Methodological morals

  • “Everyone makes statistical mistakes, not everyone fixes them.”
  • “If the numbers aren't consistent, Excel is often part of the story.”
  • “If a p-value doesn't survive the trip from text to figure, there's a problem.”

Statistical Sleuthing Extended Notes


References


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 

...
See all episodes