What is Alternative Hypothesis?

The alternative hypothesis is a statement in statistical testing that proposes a significant relationship, effect, or difference exists between two or more variables in a population. It is the direct opposite of the null hypothesis and represents the primary claim or theory that a researcher, analyst, or data scientist is actively attempting to prove with data.

 

What is the purpose of the Alternative Hypothesis?

The purpose of the alternative hypothesis is to provide a concrete claim that can be evaluated against the default assumption (the null hypothesis). It guides the direction of a statistical test. If the statistical analysis proves that the collected data is highly unlikely to occur under the null hypothesis, the alternative hypothesis is accepted as the correct explanation for the observed data.

 

How does the Alternative Hypothesis relate to the Null Hypothesis?

The alternative hypothesis and the null hypothesis are mutually exclusive.

• The null hypothesis states that any observed differences in data are strictly due to random chance.
• The alternative hypothesis states that the differences are not due to chance, but rather the result of a specific intervention or underlying relationship.

A statistical test evaluates the sample data to determine which of these two statements to accept.

 

Are there different types of Alternative Hypotheses?

Yes, an alternative hypothesis can be classified as directional or non-directional.

• A directional alternative hypothesis specifies the exact nature of the expected effect, such as stating that an intervention will specifically increase a metric.
• A non-directional alternative hypothesis simply states that an effect or difference will occur, without specifying if it will be a positive or negative change.