---
title: Getting Fit and Proving It
description: Getting Fit and Proving It
---

[Minitab Blog](https://blog.minitab.com/en/blog)

# [Getting Fit and Proving It](https://blog.minitab.com/en/blog/the-statistics-of-science/getting-fit-and-proving-it)

 Written by [Minitab Blog Editor](https://blog.minitab.com/en/blog/author/minitab-blog-editor) | Aug 16, 2012 4:00:00 AM

Now that the 2012 Olympics has concluded, the sight of all those muscled athletes has me thinking about my own physical fitness.  I’m pretty active for a 30-something.  I try to hit the gym or pool daily and I’ve been working through the [BeachBody](http://www.beachbody.com/), P90X and Insanity workout programs for the past three years. I feel pretty fit and healthy. That is why it was such a shock when I went to the doctor for my yearly physical and they gave me a report stating I was overweight. I thought, *this cannot be!*

As it turns out, my doctor’s office used the standard BMI [calculation](http://www.bmi-calculator.net/) to diagnose me as overweight. The BMI calculation uses an individual’s height and weight to determine body fat percentage. According to the National Heart Lung and Blood Institute [Web site](http://www.nhlbi.nih.gov/health/public/heart/obesity/lose_wt/risk.htm#limitations), the BMI calculation underestimates body fat in older persons who have lost muscle and overestimates body fat in those of us with a muscular build. The error rate for BMI can be as high as 20% for people with high muscle density.

## Classifying Error: Sensitivity and Specificity

This started me thinking about two common statistical terms: sensitivity and specificity.  Sensitivity measures the proportion of how many positives were positively identified. Specificity is the flip side--the number of negatives that are correctly identified as negatives. You might remember learning the [2x2 contingency table](http://support.minitab.com/en-us/minitab/17/topic-library/basic-statistics-and-graphs/tables/data-and-table-layouts/contingency-tables/) in your first statistics class and the example of guilty vs. not guilty and putting innocent people in jail. Sensitivity and specificity relate directly to that contingency table. A common tool used to identify sensitivity and specificity is [ROC Curve](http://support.minitab.com/en-us/minitab/17/macro-library/macro-files/anova-and-regression-macros/roc-receiver-operating-characteristic-curve/) graphs.   

Additionally, the concepts of Type I and Type II error can also be derived from the contingency table. A Type I error would be to identify a negative positively (also called a false positive). A Type II error would be the exact opposite: saying that a truly positive result is negative.

When we apply alpha and beta risk thresholds to our data, we are trying to account for the possibility of making the Type I and Type II errors, respectively. 

## Ultra Lean 

Since the error rate for BMI was so high for people with high muscle density, I investigated some other body fat measurement options. I decided to try the air-displacement plethysmography unit called the BodPod which had only approximately +/- 4% error rate. I worked with my dietician to schedule a BodPod appointment.  As it turns out, I was at the extreme opposite end of “overweight.”  Being a woman with just 17% body fat, I was classified as “Ultra Lean.”  Thankfully, mine was a false positive diagnosis of overweight (Type I Error). 

I guess all the blood, sweat, and tears with [Tony Horton](http://tonyhortonsworld.com/) really are paying off!

[View full post](https://blog.minitab.com/en/blog/the-statistics-of-science/getting-fit-and-proving-it)

```json
{
  "@context" : "http://schema.org",
  "@type" : "BlogPosting",
  "author" : {
    "@type" : "Person",
    "name" : "Minitab Blog Editor"
  },
  "dateModified" : "2025-07-06T00:13:55.354Z",
  "datePublished" : "2012-08-16T04:00:00Z",
  "headline" : "Getting Fit and Proving It",
  "image" : {
    "@type" : "ImageObject",
    "height" : 287,
    "url" : "https://3447555.fs1.hubspotusercontent-na1.net/hubfs/3447555/B2BML%20-%20NEW%20FILE%20MANAGER%20STRUCTURE/Blog%20Images/Imported_Blog_Media/tammys_contingency_table-3.jpg",
    "width" : 452
  },
  "mainEntityOfPage" : "https://blog.minitab.com/en/blog/the-statistics-of-science/getting-fit-and-proving-it",
  "publisher" : {
    "@type" : "Organization",
    "logo" : {
      "@type" : "ImageObject",
      "height" : 60,
      "url" : "/hs/hsstatic/content_shared_assets/static-1.4092/img/default-amp-logo.png",
      "width" : 60
    },
    "name" : "Minitab Blog"
  }
}
```