Question

In which of the following situations would you use a one-tailed test (as opposed to a twotailed test)?

A. When you are testing only one group of people, rather than two

B. When there is a clear reason to predict a result in one specific direction

C. When you are using the normal distribution instead of the T distribution

D. When you want to avoid a Type 1 error

Answer 1

In one sample test, suppose researcher wants to know population mean is less than Or grater than of specific population mean. Using sample observations we can do this type of test. (Same procedure is applied in proportion test). In one tailed test researchers has clear reason to predict the result in one specific direction instead of two direction.

(Two direction means that alternative hypothesis is like population mean is different from specific population mean. That is it can be less than Or grater than type. We cannot predict it is less than type or grater than type.).

In one tailed test researcher can predict the alternative hypothesis in one specific direction.

Therefore answer is,

B. When there is a clear reason to predict a result in one specific direction.