Survey Tabulation > Statistical tests > p values
 
p values
The p (probability) value is the basis for deciding whether or not there is a relationship between the data being tested. Generally you start with the null hypothesis, which is the assumption that there is no relationship between the data. If the p value is small enough (usually less than 0.05 or 0.01), you can reject the assumption of no relationship and conclude that there is a relationship.
A full list of the p values for each pair of columns tested in each row is available in the diagnostics information file for the table. UNICOM Intelligence Reporter - Survey Tabulation reports p values as a decimal value with six decimal places (although trailing zeroes after the decimal symbol are not displayed by default when you open the file in Microsoft Excel). A reported p value of 0.05 is equal to a significance level of 5%. If the p value is smaller than the significance level you select for the test, the test statistic is significant.
See Diagnostics information for further details.
The p value is also called the observed significance level.
p values for the column proportions test example
 
rows/cols
pval
Dinosaurs
1AB
0.132292
Conservation
2AB
0.717116
Fish and reptiles
3AB
0.107254
Fossils
4AB
0.545771
Birds
5AB
0.453826
Insects
6AB
0.102733
Whales
7AB
0.941864
Mammals
8AB
0.717116
Minerals
9AB
0.766192
Ecology
10AB
0.978717
Botany
11AB
0.486575
Origin of species
12AB
0.169406
Human biology
13AB
0.009396
Evolution
14AB
0.606386
Wildlife in danger
15AB
0.813749
Other
16AB
0.935494
Not answered
17AB
0.211844
This example shows the p values for the first column proportions test example. It shows the row/cols and pval columns from the diagnostics data with the addition of the row labels. In the example, the column proportions test was run at the 5% significance level, and one row, Human biology, was found to be significant. Notice that the p value in this row is 0.009396. This is the only row with a p value of less than 0.05.
See Examples of the column proportions test for more information.
See also
Statistical tests