Understanding the Root Mean Square (RMS) and Root Sum of Squares (RSS): A Practical Guide for Medical Laboratory Scientists

In medical laboratories, uncertainty analysis is crucial for reliable results. Root Mean Square (RMS) estimates the quadratic mean of repeated measurements, while Root Sum of Squares (RSS) combines uncertainty from multiple sources. RMS is for describing repeated measurements, while RSS is for combining different sources of uncertainty. Both methods are essential for managing measurement uncertainty.

In medical laboratories, precision is everything. But how do we account for the variability in measurements and calculate the overall uncertainty? Uncertainty analysis in medical laboratories is a crucial part of ensuring reliable results, and two key methods—Root Mean Square (RMS) and Root Sum of Squares (RSS)—are often used to describe data for uncertainty.

While these methods may sound similar, they serve very different purposes. This guide will help you understand Root Mean Square in lab testing and Root Sum of Squares calculation, providing examples to clarify when and how to use each method for measurement uncertainty in laboratories.

Understanding Uncertainty Analysis in Medical Laboratories

When we measure something multiple times in a lab, the results don’t always match exactly. So how do we summarise the mean value in a different way to the traditional arithmetic mean, or combine different measurements? This is where RMS and RSS come in.

Both techniques involve square-rooting, but beyond that, they’re quite different in how they help you handle variability.

Root Mean Square in Laboratory Testing: How It Works

Imagine you’re measuring a patient’s glucose levels multiple times in a row. Each result may vary slightly due to small factors like temperature or analytical imprecision. The Root Mean Square (RMS) is a useful tool for calculating the average in those repeated measurements.

A Simple Analogy: Rolling Dice

Think of rolling dice multiple times. You’ll get different results each time, even though the dice themselves haven’t changed. This is similar to running the same lab test repeatedly—there’s always some variation. The RMS helps estimate how big those variations are, on average.

Example of RMS Calculation in Lab Testing

Let’s say you’ve measured the glucose level of a patient five times:

• 99 mg/dL
• 101 mg/dL
• 100 mg/dL
• 98 mg/dL
• 102 mg/dL

To calculate the RMS:

Square each measurement:

• 99² = 9,801
• 101² = 10,201
• 100² = 10,000
• 98² = 9,604
• 102² = 10,404

Sum these squared values

9,801 + 10,201 + 10,000 + 9,604 + 10,404 = 50,010

Divide by the number of measurements (N)

50,010​/5=10,002

Take the square root

≈100.01

The RMS tells you that, on average, your measurements are around 100.1 mg/dL.

The Root Mean Square Error (RMSE) defines the differences between values predicted by model (such as a mean, or a regression) and the actual values. It is the standard deviation of the differences between predicted and observed values. Commonly referred to as residuals they are considered as error in the prediction of the model – a measure of how good the model fits the underlying data.

• 99 – 100.1 = -1.1
• 101 – 100.1 = 0.9
• 100 – 100.1 = 0.1
• 98 – 100.1 = -2.1
• 102 – 100.1 = 1.9

We then add up all the differences

-1.1 + 0.9 + 0.1 + -2.1 + 1.9

= -0.2

divide by the number of samples, then take the square root

RMSE = 0.09 so +/- 0.1 at a level of 100.1 (keeping significant figures the same between the uncertainty and the measured value (RMS in this case).

Root Sum of Squares: Combining Measurement Uncertainty

Let’s say you have uncertainty from multiple sources—like calibration, imprecision, and bias correction. To combine these sources into a single measure of uncertainty, you use Root Sum of Squares (RSS).

A Simple Analogy: The Pythagorean Theorem

Just as the hypotenuse of a right triangle is the square root of the sum of the squares of the other sides, RSS works by summing squared uncertainties from different sources and then taking the square root of the result.

Example of RSS Calculation in Lab Testing

Consider three sources of uncertainty in a glucose test:

• Calibration: ± 1 mg/dL
• Intermediate precision: ± 2 mg/dL
• Bias correction: ± 1.5 mg/dL

To combine these uncertainties:

Square each uncertainty contributor

• Machine calibration: 1² = 1
• Reagent precision: 2² = 4
• Operator variability: 1.5² = 2.25

Sum the squared values

1 + 4 + 2.25 = 7.25

Take the square root

≈2.69

Your combined uncertainty is ± 2.69 mg/dL. This combined uncertainty is less than the sum of all individual uncertainties, ensuring that the total is accurate without overestimating it.

Conclusion: Tools for Managing Uncertainty in Laboratories

Understanding Root Mean Square (RMS) and Root Sum of Squares (RSS) helps you manage measurement uncertainty in laboratories. Both methods are essential tools, but they serve different purposes:

• Use RMS when dealing with repeated measurements of the same thing.
• Use RSS when combining multiple sources of uncertainty into a single estimate.

By applying these techniques, you can ensure your lab’s measurements are both accurate and reliable.

Want to learn more try out these courses!

If you enjoyed this post, explore more on our online learning platform. We offer comprehensive online courses designed to enhance your knowledge and skills. Click here to check out our courses and continue your learning journey with us! Below is a free course to get you started with descriptive statistics in Excel