2.4 SPSS and Theoretical Approximation of the Sampling Distribution

By default, SPSS uses a theoretical probability distribution to approximate the sampling distribution. It chooses the correct theoretical distribution but you yourself should check if the conditions for using this distribution are met. For example, is the sample large enough or is it plausible that the variable is normally distributed in the population?

In one case, SPSS automatically selects an exact approach if the conditions for a theoretical approximation are not met. If you apply a chi-squared test to a contingency table in SPSS, SPSS will automatically apply Fisher’s exact test if the table has two rows and two columns. In all other cases, you have to select bootstrapping or an exact approach yourself if the conditions for a theoretical approximation are not met.

We are not going to practice with theoretical approximations in SPSS, now. Because theoretical approximation is the default approach in SPSS, we will encounter it in the exercises in later chapters.