Hypothesis Testing

Hypothesis testing is when you take a hypothesis (an educated guess about something you can test) and then use statistics to determine the probability that the hypothesis is true.

Hypothesis testing is one of those things we take for granted, because it’s used so often in so many areas, yet it’s important to remember there is an official recipe for it, and that if something’s not proven true, it doesn’t mean it was proven false.

The four hypothesis testing steps are:

1) Make a hypothesis (oftentimes, this is guessing that what we observe is due to random chance, called the “null hypothesis”) and an alternative hypothesis (oftentimes this is a guess as to a cause-and-effect relationship, in addition to other factors and random chance).

2) Find a test statistic that can be used to assess the truth of the null hypothesis (your testing tool).

3) Calculate the p-value, which will determine statistical significance.

4) Compare the p-value to the tested data to see if it’s statistically significant, which means you’re on to something...probably something really, really specific.