agulani

The basic premise of hypothesis testing is that you do not know the value of a parameter of a population. You want to test that it is not some value, say 0 in your example above. You test it by taking a sample of values and finding their mean. Depending on what you get, you can say whether the population mean has a 95% probability of being 0 or 90% depending on significance level chosen. (THink of it this way, if sample mean is very close to 0, there is a good chance that pop mean is 0).

Even if pop mean is 0, the sample mean you get may not be 0 since it doesnt include all members of the population. But it may be so small that it is not significantly different from 0 (statistically speaking)







You think that the population mean is something other than 0 so you state
Ho - mean = 0
Ha - mean not equal to 0