Definition
A Type 1 error (also called a false positive or alpha error) occurs when you reject a null hypothesis that is actually true. You conclude there is an effect or difference when none really exists.
How It Happens
Even with good data and correct procedures, random chance can produce results that look significant. The significance level (alpha) is the maximum probability of making this mistake that you are willing to accept.
A food company tests whether a new recipe tastes better than the original.
In reality, customers have no preference. But by chance, the test sample slightly favors the new recipe, producing p = 0.04.
The company concludes the new recipe is better and launches it. This is a Type 1 error - they "found" a difference that does not actually exist.
Why It Matters
Type 1 errors lead to false discoveries, wasted resources, and wrong decisions. In medicine, a Type 1 error might mean approving a drug that does not actually work. In business, it might mean investing in a strategy based on a fluke result.
The significance level of 0.05 means you accept a 5% chance of a Type 1 error on each test. When running many tests, these chances accumulate. If you test 20 things, you expect about one false positive even if nothing is real. This is why corrections for multiple comparisons (like Bonferroni) exist.
A Type 1 error means finding an effect that is not there. The significance level controls this risk, but multiple testing can inflate it. Always consider replication.