卡方检验(国外英文资料).docVIP

卡方检验(国外英文资料) Chapter 16 statistical inference of the disorder classification variable -- the card square test Through introducing the front, can know can be divided into continuous variables (distance, constant ratio) and classification variables, which in turn is subdivided into orderly, two kinds of chaotic variables. Quantitative variables for each groups overall average (i.e., continuous variables), you can use the t test and variance analysis method, rank and inspection is used for comparison between groups in general for the distribution of orderly classification variables are the same. Here is the card party inspection is mainly used for statistical inference, disorderly classification variables is can t test in the application of degree of another commonly used inspection method. Continuous variable t test Multiple variance analysis The classification variable is ordered rank and test Unordered card testing 16.1 caside inspection overview The basic principle of the test of 16.1.1 The basic idea of the caside test The card square test is a commonly assumed test method based on the chi 2 distribution, Its null hypothesis is H0: there is no difference between the observed frequency and the expected frequency. The basic idea of the caside test is that, starting with H0, the chi 2 value is calculated based on this premise, which indicates the degree of deviation between the observed value and the theoretical value. According to the chi 2 distribution and the degree of freedom, the probability of obtaining the current statistic and the more extreme situation is given by the fact that the H0 hypothesis is true. If the value of the P is small, the deviation of the observed value and the theoretical value is too great, and the original hypothesis should be rejected, indicating that there is a significant difference between the comparison data. Otherwise, we cant reject the null hypothesis, but we cant say that the actual situation that the sample represents is differ