Confirming that the data is random – the null hypothesis
One of the important statistical questions is framed as the null hypothesis and an alternate hypothesis about sets of data. Let's assume we have two sets of data, S1 and S2. We can form two kinds of hypothesis in relation to the data:
- Null: Any differences are random effects and there are no significant differences.
- Alternate: The differences are statistically significant. Generally, we consider that the likelihood of this happening stochastically to samples that only differ due to random effects must be less than 5% for us to deem the difference "statistically significant."
This recipe will show one of many ways in which to evaluate data to see whether it's truly random or whether there's some meaningful variation.
Getting ready
The rare individual with a strong background in statistics can leverage statistical theory to evaluate the standard deviations...
