Summary
Highlights
The lesson introduces P-values in the context of large sample hypothesis testing of means. It highlights that while rejection regions are valid, P-values are more common in actual research and provide an alternative method for deciding whether to reject a null hypothesis.
A P-value is defined as the probability of obtaining a sample more extreme than the ones observed in your data, assuming the null hypothesis is true. The term "more extreme" depends on whether it's a left-tail, right-tail, or two-tail test.
For a left-tail test, the P-value is the area under the Z-distribution curve to the left of the calculated test statistic (Z-value). This area represents the probability of observing data as or more extreme than the collected sample data in the left direction.
In a right-tail test, the P-value is the area under the Z-distribution curve to the right of the calculated test statistic (Z-value). This signifies the probability of observing data as or more extreme than the collected sample data in the right direction.
An example demonstrates calculating the P-value for a left-tail test where the alternate hypothesis suggests a mean less than 0.15. Given a test statistic Z = -1.34, the P-value is found by looking up the area to the left of -1.34 in a Z-distribution table, yielding a P-value of 0.091.
This section provides an example for a right-tail test, where the alternate hypothesis suggests a mean greater than 0.43. With a test statistic Z = 2.78, the P-value is the area to the right of 2.78. Using the Z-table, which typically gives the area to the left, the calculation involves finding the area to the left of -2.78 due to symmetry, resulting in a P-value of 0.0027.