A sign test (also known as a two-sample sign test, or the binomial sign test) is a type of comparison test used to compare paired data, using + and - signs to describe the direction of differences between pairs.
This test is non-parametric, which means that it does not depend on the data having a normal distribution in order to produce results that researchers can believe. In this way, sign tests are suitable to be used with small samples. Therefore, it can be thought of that the sign test is the non-parametric equivalent to the paired t-test.
The null hypothesis H_0 of the sign test is that the medians of both groups are equal to each other, and the alternative hypothesis H_A is that the medians are not equal to each other.
The sign test is non-parametric, which means it does not require any assumption about the data's distribution. However, there are still some other assumptions which must be met:
If your data meets these assumptions, you can perform the sign test.
A special case of the sign test is the one-sample sign test, which is a test used to investigate whether the median of a single sample is statistically different to a single value. In this sense, the one-sample sign test is very similar to the Wilcoxon signed-rank test. Indeed, the one-sample sign test is the non-parametric equivalent to the one-sample t-test.
The one-sample sign test has one less assumption to the sign test, in that it does not assume that data needs to be matched (or paired).
The null hypothesis H_0 of the one-sample sign test is that the median of the sample is equal to the known value, and the alternative hypothesis H_A is that the median is not equal to the known value.
In SPSS, lay out your data so that your two groups are two variables. Make sure that your data is paired, so that each participant's results are in the same row.
When you are ready to perform the test:
Your output will consist of three tables.
