type I error

n. Type I error is the likelihood of falsely concluding that a trend in sample data is actually present in the population at large. Sometimes referred to as making a “false alarm,” type I errors are sometimes very costly, depending on the situation and the implications to be drawn from the research findings. Traditionally, the statistical cutoff point for making a type I error is a probability of 5% or less that a trend in the data occurred by chance alone, though researchers may choose a more conservative decision point (or alpha level) if the cost of type I errors is high.  – MWP

▶ See also P-VALUE, SIGNIFICANCE TEST, and STATISTICAL SIGNIFICANCE