Principles of Statistics in Respiratory Research: Part 6. Hypothesis Testing

The steps involved in hypothesis testing are to determine hypotheses, to choose and compute the test statistic, to determine the rejection region for the test statistic, and to make a conclusion. The choice of the statistical test depends on the type of response variable, the number of factor levels, the type of design, and the size of the sample. After computing the test statistic, the researcher makes one of two possible conclusions and is concerned with an erroneous conclusion. The two possible types of errors are called Type I and Type II errors. Finally, when more than one statistical test is appropriate, the tests can be compared by their power. Various statistical tests for some commonly used designs are presented in this article. The important tasks for the researcher are to choose the appropriate test, to understand the hypothesis being tested, and to interpret the results.