Rejecting a true null hypothesis — a false discovery.A Type I error, also known as a false positive, occurs when a statistical test rejects a true null hypothesis. In plain terms, it means concluding that an effect or difference exists when in reality it does not. This is the error that scientists most often guard against, because it leads to claiming discoveries that are not real.
The formal concept was developed in the early 20th century by statisticians such as Ronald Fisher and Jerzy Neyman, with Neyman and Egon Pearson explicitly distinguishing Type I and Type II errors in the 1920s–1930s.
In hypothesis testing, you start with a null hypothesis (e.g., 'the drug has no effect') and an alternative hypothesis (e.g., 'the drug has an effect'). You collect data and compute a test statistic. If the statistic falls into a region of extreme values (the rejection region) that would be unlikely if the null hypothesis were true, you reject the null. The probability of this happening by chance is the significance level, typically set at 0.05. A Type I error is exactly this: rejecting the null when it is actually true, and the significance level is the probability you allow for that mistake.
Used in every field that relies on statistical inference—medicine (drug efficacy trials), psychology, economics, and quality control—to set thresholds for what counts as a 'significant' result.
The standard 5% significance level means that if you test 20 truly ineffective drugs, on average one will appear effective purely by chance.
