1. statistical research design: the name and main practices of statistical research design should be explained. Such as survey design (divided into prospective, retrospective or cross-sectional survey research), experimental design (specific design types should be explained, such as self-paired design, grouping design, crossover design, factorial design, orthogonal design, etc. ), clinical trial design (which stage of clinical trial should be explained, which blind measures should be adopted, etc. ); The main methods should be outlined around four basic principles (repetition, randomness, contrast and balance), especially how to control the interference and influence of important non-experimental factors.

2. Representation and description of data: X S is used to represent quantitative data that approximately obeys normal distribution, and M (QR) is used to represent quantitative data that is skewed distribution; When using statistical tables, it is necessary to reasonably arrange vertical and horizontal headings and clearly express the meaning of data; When using statistical charts, the types of statistical charts used should match the data properties, and the marking method of scale values on the number axis should conform to mathematical principles; When using relative numbers, the denominator should not be less than 20, and attention should be paid to distinguishing percentages from percentages.

3. Selection of statistical analysis methods: For quantitative data, appropriate statistical analysis methods should be selected according to the design type, conditions and analysis purpose of the data, and T test and one-way ANOVA should not be blindly applied; For qualitative data, appropriate statistical analysis methods should be selected according to the design type, the nature and frequency of qualitative variables and the purpose of analysis, and χ2 test should not be blindly applied. For regression analysis, we should combine professional knowledge and scatter plot to choose the appropriate regression type, and we should not blindly apply simple linear regression analysis and simplify the processing of regression analysis data with repeated experimental data. For multi-factor and multi-index data, on the basis of univariate analysis, multivariate statistical analysis method should be used as much as possible, so as to make a comprehensive and reasonable explanation and evaluation of the interaction between factors and the internal relationship between multi-indicators.

4. Interpretation and expression of statistical results: When P