Factor analysis is a commonly used data analysis method in academic papers, which refers to the statistical technique of extracting * * * sexual factors from variable groups. Factor analysis is to find internal relations from a large number of data and reduce the difficulty of decision-making.

Common functions of factor analysis

(1) In regression analysis, solve * * * linear problem: If there is * * * linear problem in regression analysis, a representative common factor can be extracted from multiple variables with * * * linear problem, and the extracted common factor can be used to replace multiple variables with * * * linear problem and participate in modeling, so that the * * * linear problem in regression analysis can be solved.

(2) Variable simplification: Generally speaking, the fewer variables contained in the model, the better. If there are many variables, we can simplify the variables by extracting common factors first, so that the variable information contained in the model is not greatly attenuated and the complexity of the model is also reduced.

(3) Validity analysis in the questionnaire: For the scale questions in the questionnaire, I hope to find the questionnaire structure through factor analysis, test the structural validity of the questionnaire, and divide the scale questions into different scoring dimensions according to factor analysis.

3. Regression analysis

In the data analysis methods commonly used in academic papers, the influence of a random variable Y on another (x) or a group (X 1, X2,? , Xk) statistical analysis method of variable correlation. Regression analysis is a statistical analysis method to determine the quantitative relationship between two or more variables.

Regression analysis classification

(1) one-dimensional linear regression analysis

Only one independent variable X is related to the dependent variable Y. Both X and Y must be continuous variables, and the dependent variable Y or its residual must obey normal distribution.

(2) Multiple linear regression analysis

Conditions for use of multivariate linear regression analysis: To analyze the relationship between multiple independent variables and dependent variable Y, both X and Y must be continuous variables, and dependent variable Y or its residuals must obey normal distribution.

(3) Logistic regression analysis

Linear regression model requires the dependent variable to be continuously normal distribution, and the independent variable has a linear relationship with the dependent variable, while Logistic regression model does not require the distribution of the dependent variable, and is generally used when the dependent variable is discrete.

(4) Other regression methods

Nonlinear regression, ordered regression, probability unit regression, weighted regression, etc. Because there are many types of regression analysis, it is very important to choose the specific regression type according to the dimension of data and other basic characteristics of data when choosing regression methods, which is very important for the later data analysis.

4. Analysis of variance

Used to test the significance of the difference between two or more samples. Due to the influence of various factors, the data obtained from the research is fluctuating. The causes of fluctuations can be divided into two categories, one is uncontrollable random factors, and the other is controllable factors that affect the results. Analysis of variance starts with the variance of observed variables, and studies which variables among many control variables have significant influence on observed variables.