I. Descriptive statistics

Descriptive statistics is a method of sorting out and analyzing data through charts or mathematical methods, and estimating and describing the relationship between data distribution, digital characteristics and random variables. Descriptive statistics can be divided into three parts: centralized trend analysis, decentralized trend analysis and correlation analysis.

Second, correlation analysis.

Correlation analysis is a statistical analysis method to study the correlation between two or more random variables in the same position. For example, between a person's height and weight; The correlation between relative humidity in the air and rainfall is a correlation analysis problem.

1, single correlation: refers to the correlation between two variables. Such as the relationship between product output and unit product cost. There is only one dependent variable and one independent variable.

2. Multiple correlation: refers to the correlation between one variable and two or more other variables.

3. Partial correlation: when a phenomenon is related to multiple phenomena, two random variables are called partial correlation when the influence of other random variables is excluded.

Third, analysis of variance.

By analyzing the contribution of variation from different sources to the total variation, we can determine the influence of controllable factors on the research results. The research sources must be independent of each other and the total variance is equal.

1. One-way ANOVA: When there is only one influencing factor or multiple influencing factors in the study, only the relationship between one factor and the response variable is analyzed.

2. Multi-factor analysis of variance: there are two or more factors affecting the dependent variable, and the relationship between multiple factors is considered at the same time.

3. Multi-factor non-interactive variance analysis: analyze the relationship between multiple factors and dependent variables, but there is no influence relationship between them or the influence relationship is ignored.

Fourth, hypothesis testing.

1, parameter test: Its basic principle is to test some main parameters when the overall characteristics are known.

2. nonparametric test: nonparametric test is a method to infer the general distribution pattern by using sample data when population variance is unknown or little known. The main methods are: chi-square test of population distribution, binomial distribution test, single sample K-S test and so on.