The null hypothesis is the complement of the alternative hypothesis. The null hypothesis is what you anticipate through randomness. The alternative hypothesis, sometimes known as the alternate hypothesis, is the opposite of that. It is what you would not anticipate by randomness.
In other words, the alternative hypothesis “rejects the null” and shows evidence of correlation between two sets of data.
In Six Sigma, much of the statistical analysis seeks to reject the null hypothesis. It looks to find relationships between variables and identify what action (or lack of action) causes unwanted variation and errors. In this way, teams can then focus on the root cause of a problem.
Steps For Hypothesis Testing
The steps for a basic hypothesis testing involve identifying the question, determining the significance, choosing the test, interpreting the results and making a decision. In Six Sigma, the focus is typically on rejecting the null and finding correlation between variables.
Creating a null hypothesis typically involves taking a question and rephrasing it as a statement.
Q: Do customers order more appetizers if they’re enlisted on the first page of the online menu?
Null: Placement of the appetizer list has no impact on how often people order appetizers.
Q: Do website visits click on the “find out more” button if it is blue rather than red?
Null: The color of the “find out more” button has no impact on how often people click on it.
Sometimes it’s best to start off by writing down the alternative hypothesis, known as the alternate hypothesis statement. For example, the statement might be that method A will produce a more absorbent product than method B. With this statement, the team is saying that when they measure the characteristic of some material to be more absorbent, method A will hopefully be found to be a larger-is-better characteristic as compared to method B.
Then, writing the null hypothesis is simply stating the opposite of the alternative hypothesis. In this case, the null hypothesis would be that there is no statistically significant difference in using method A vs. method B.
A null hypothesis provides a starting place to determine basic business issues such as the impact of different materials on the quality of the final product, the types of advertising that lead to better sales or the impact of social media postings on website traffic.
Null hypothesis also helps investors make financial decisions. For example, they will look at return rates of mutual funds over many years to determine what variables do or do not have an impact on annual returns.
