How to Read a Chi-Square Chart
- 1). Write your calculated degrees of freedom (df). For example, you write "3." Gerard E. Dallal, Ph.D., calls degrees of freedom "a way of keeping score." It is the number of categories in your study minus 1. For example, if your study involves two categories, yes or no, your degree of freedom is 1 (2 - 1 = 1).
- 2). Write the p-value you will use as your standard for accepting or rejecting the results of your study. Biologists often use a p-value greater than 0.05 (p > 0.05). You write, "p > 0.05."
- 3). Write your calculated chi-square value. For example, you write, "1.411." A researcher calculates his chi-square value by subtracting each expected result in his study from each observed (or actual) result, squaring the answer, dividing by the expected result, then doing this for every observed result, and adding all the answers.
- 4). Refer to a chi-square distribution table, or chart, in a textbook or online.
- 5). Locate your degrees of freedom value in the column on the far left-hand side. For example, you locate the value 3 in the column on the far left-hand side. It is in the fourth row from the top.
- 6). Move right along the degrees of freedom value row until you reach the number that is closest to your calculated chi-square value. For example, you move four columns to the right until you reach the number 1.42. This is the number in the row closest to your chi-square value.
- 7). Read up the column where you have found the closest number to your chi-square value to the top number in the column. This is the p-value associated with your chi-square value. For example, you read up the column to 0.70.
- 8). If this value is greater than the standard of probability you have decided to use, accept your hypothesis. For example, 0.70 is greater than 0.05. You accept your hypothesis.
- 9). If this value is less than the standard of probability you have decided to use, reject your hypothesis. For example, if your chi-square value were 11.37 with three degrees of freedom, your p-value would be 0.01. The p-value 0.01 is less than 0.05. You must reject your hypothesis.
