Abstract
Count data are very common in biomedical research, and such kind of data is often organized in the form of contingency table. For count data, the common research purpose is to test whether two factors are independent, therefore chi-square test is often used for statistical analysis, but it is easy to overlook the applicable conditions of chi-square test and the correction of results under different conditions. In addition, for count data, there are also other research purposes, such as testing whether there is a linear trend between two categorical variables, whether the results are consistent, and so on. Therefore, how to choose appropriate testing methods based on different research purposes is also a problem worth paying attention to in practical applications. In this study, the commonly used statistical methods for various count data are systematically summarized through a series of examples, and the common misuses of the chi-square test are analyzed.