ABSTRACT
Statistical Guide: A one-way chi square (also known as a goodness-of-fit chi square) tests for the significance of the difference(s) among frequencies when subjects are classified in only one way. For example, we might ask a random sample of subjects for whom they plan to vote. If 60 say Candidate A and 30 say Candidate B, the null hypothesis asserts that the difference is due to the sampling errors. If a chi square test yields a p value of .05 or less, we usually reject the null hypothesis and declare the difference to be statistically significant. Note that in this example subjects are classified only according to whom they plan to vote for-they are classified in only one way. If we also classified them in terms of gender (i.e., how many men and how many women plan to vote for each candidate), we could conduct a two-way chi square, which is taken up in Exercise 32.
