t-Tests
Statistics

Single Sample t-Test

This test evaluates whether a sample of n observations come from a parent population in which the mean equals a specific (hypothesized) value.

  • The data should be stored in a single numeric column.

    Example: Consider a numeric column of data that represents the content of organic carbon in a given rock formation from different cores. We hypothesize that the parent population mean value is 5. Running a single sample t-test will examine this hypothesis. (see the image left-hand side image below).

Paired Samples t-Test

This test evaluates whether two dependent samples represent populations with different means. The two samples must have the same number of data points, and must be matched as nearly as possible to exclude the effects of extraneous variation. Examples are two variables measured before and after an intervention

  • The data on the dependent samples should be stored in two separate numeric columns.

    Example: Consider two numeric data columns representing a pair of measurements taken on the same subjects at two different times.

Unpaired Samples t-Test

This test evaluates whether two independent samples represent two populations with different means, while assuming the populations having the same variances. The two samples/groups can have the same or a different number of data points.