A two-sided hypothesis is used when there is no strong directional hypothesis expected in a research study. Essentially, the researchers do not know if a treatment effect will cause an increase or decrease in some sort of outcome.
A two-sided hypothesis requires a relative large difference between groups, observations, or association in order to detect significance versus a one-tailed test. Two-tailed tests are therefore thought to be more rigorous because more observations have to be collected in order to achieve adequate statistical power.
A two-sided hypothesis requires a relative large difference between groups, observations, or association in order to detect significance versus a one-tailed test. Two-tailed tests are therefore thought to be more rigorous because more observations have to be collected in order to achieve adequate statistical power.
Click on the Statistics button to continue.