alternative hypothesis

n. In statistics, the hypothesis that there is no difference between the means of different groups in an analysis of variance, implying that there is no relationship between levels of the independent variable and the dependent variable. This is used as a dummy hypothesis so that when it can be statistically rejected, the null hypothesis that there is a difference between groups can be accepted and so demonstrate that it is not unreasonable to believe that there has been an effect produced by the independent variable on the dependent variable.