In: Economics
QUESTION 4
In which case would testing the null hypothesis not involve a one-tailed statistical test? PLEASE PROVIDE BRIEF EXPLANATION!
H1: Incentive pay for teachers does affect student achievement |
||
H1: Higher sales tax rates does not reduce state tax revenues |
||
H1: Extending the duration of unemployment benefits does not increase the length of joblessness |
||
H1: Smoking does not reduce life expectancy |
In two-tailed hypothesis testing, we check the possibility of relationship being affected in both direction.
For example, our null hypothesis can be - Humidity does not effect vehicle speed. The alternate hypothesis will be humidity does effect vehicle speed. (It could decrease or increase the speed)
One tailed test on the other hand only checks increase or decrease (and not both).
For example, we have discovered a new medicine which costs less than the already available one. Since it is cheaper, we just need to ensure the new medicine is not less effective than the existing one. (If it is not less effective than existing one, we will replace the existing medicine). In this case, we use one-tailed test.
In this example, all options except the first one is only checking in one direction and therefore, one-tailed test may be used. But for the first option, (Incentive pay for teachers does affect student achievment), ncentive pay could increase r decrease student achievement. Hence we need two-tailed test for this. Hence first option is the answer.