Why calculating the difference between two measurements can lead to misleading scientific results
In short: Researchers who use longitudinal data often calculate 'change scores' by subtracting a baseline measurement from a follow-up result to see how much a variable shifted. However, a study using simulations and mathematical models found that this common practice can produce inaccurate estimates of cause and effect in observational studies.
The study used directed acyclic graphs and simulations to test three scenarios: when the baseline measurement was a competing exposure, a confounder, or a mediator. The results showed that change scores only provide meaningful causal estimates in the first scenario.
In cases where the baseline measurement acts as a confounder or a mediator, change scores can produce results that differ significantly in size and direction from the actual causal effects. Because the study relied on simulations rather than human subjects, it demonstrates a mathematical principle rather than a real-world clinical outcome.
Why it matters: Simple subtraction of measurements can hide the true relationship between variables, potentially leading to incorrect scientific conclusions.
Opens the Intellect+ feed with this card first. Swipe up for the next one.
Sources
To cite this card, link to https://intellectplus.co.uk/app/science-research-why-calculating-the-difference-between-two-measurements-can.
How we source, check and correct cards: Editorial standards.
General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.
Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?
Related
More in Science · Latest · A–Z · All topics
