Significance test enables you to determine if two data points are significantly different from each other. Suppose we have a group of 2000 persons participating in an IQ test. After doing the test, we find out that the mean IQ of women is two percent higher than men’s. This may seem trivial but using the significance test, we can say that this difference is statistically significant because the sample size is rather large.
Significance test is used in surveys. For behavioral data, we use the 68-95-99.7 rule. In this method, we assume that the data is normally distributed. Then we calculate the mean and standard deviation values and report the points that lie outside the areas of three standards deviations of the mean as outliers.