How to Find p-Value in Research

When a researcher is about to test a hypothesis regarding a population, he can make use of the test statistic in order to determine if the null hypothesis is to be rejected or accepted. This decision is made by determining a number which is known as the p-value.

Determining p-Value in Null Hypothesis Testing

A probability that is linked with the critical value is known as the p-value. This critical value is based upon the probability for the Type I error. It calculates the probability of getting outcomes as strong as yours in the case where the null hypothesis holds true. It should be noted however that in the case where the alternative hypothesis is less-than alternative, the null hypothesis is rejected only if on the distribution the test statistic is observed to be on the left part of the tail which means below -2. In the same way, if the alternative hypothesis is more than the alternative, then the null hypothesis is rejected in the case where the distribution has the test statistic falling on the right side which means above 2.

Finding p-Value for the Test Statistic

Following are the ways in which the p-value can be found for a test statistic:

• On the right type of distribution find your test statistic. An example of this can be the standard normal distribution.
• After this you need to determine the probability that standard normal distribution is above your test statistic.
• If the alternative hypothesis has a less-than alternative, then the probability where the normal distribution is less than the test statistic is to be determined. This is basically the p-value. In this particular case, the test statistic is mostly negative.
• In the case where the alternative hypothesis has a more than alternative, the probability where the normal distribution is more than the test statistic is to be determined. The outcome of this is the p-value again. However, in this situation, the test statistic is mostly positive.

Cases for Not-Equal-To Alternative Hypothesis

In the case where the alternative hypothesis has a not-equal-to alternative, the probability where the normal distribution is beyond the test statistic is determined and then doubled. This has two situations or cases:

• Firstly, if the test statistic appears to be negative, the probability where the normal distribution is less than the test statistic is to be determined and doubled to achieve the p-value.
• Secondly, if the test statistic appears to be positive, the probability where the normal distribution is more than the test statistic is to be determined and the outcome doubled to achieve the p-value.