Developing Statistical Hypotheses
We always have a null hypothesis (H0) and an alternative hypothesis H1).
We do statistical tests to decide which of the two hypotheses may be correct.
We always start by assuming the null hypothesis H0 is correct.
After carrying out the statistical tests, we either:
- Reject H0 in favour of H1
- OR Fail to reject H0 (NOTE: this is different from saying we “accept H0”)
Worked Example
A comparison of the temperature of sea water was made at two locations. Over a 2-month period, daily readings were taken at each location. We wish to test whether the average temperature at the two locations is different. What would be our testing hypotheses?
We would test the following hypotheses: NOTES: (a) μA means the mean (average) of the readings from location A. (b) This is referred to as a two-tailed test as we are testing for μB being either larger and smaller than μA. |
A comparison of the temperature of sea water was made at two locations. Over a 2-month period, daily readings were taken at each location. We wish to test whether the average temperature at location A is warmer than at location B. Which of the following would be our hypotheses in this case?
NOTE: This is called a one-tail test because we are only testing whether location A is warmer than location B (not warmer OR colder).