Regression testing in statistics allows you to find the relationship between two quantitative variables and determine whether the trend observed is of statistical significance.
If we want to determine whether the size of the car is related to the fuel efficiency of the car, we would use a regression test.
Step 1: Hypothesis
- H0 is our null hypothesis i.e. the difference between what we observe and what we expected to observe is null and is a result of normal fluctuations.
- There is no statistical significant between the mass of the car and the fuel burnt per 100 km
- H1 is our alternative hypothesis and assumes that the differences between what we observed and what we expected is of statistical significance.
- There is statistical significant between the mass of the car and the fuel burnt per 100 km
Step 2: Analysis
- The data should have a linear trend
- There should be no relationships within the residuals
Note we use n – 2 for our degrees of freedom where n is the number of cars we have observed.
Find the test statistic and examine the value of p
Step 3: Draw a conclussion
- So reject H0 if our p value is less then the significance level
- Fail to reject H0 if our p value is greater then the significance level