Factor analysis (exploratory factor analysis) is used to explore how many factors (variables) the analysis items (quantitative data) should be divided into. For example, it is more appropriate to divide 20 scale items into several aspects; Users can set the number of factors themselves. If not, the system will set the number of factors whose feature root value is greater than 1 as the judgment standard. Select the factor method in the advanced method module, drag the analysis items to the analysis box on the right, and click Start Analysis.

Supplementary note: If you want to extract the number of factors, you can actively set the number of factors to be output. Ticking "factor score" and "comprehensive score" will generate new variables in the analysis box on the left, such as CompScore***** (comprehensive score) and Factorscore * * * * * (factor score). Factor scores can be used for further analysis, such as cluster analysis and regression analysis. Comprehensive scores can be used to compare rankings and so on.

Number of factors: in most cases, we will have subjective expectations when analyzing, and we hope to classify the items. At this time, we can directly set the corresponding factor number.

Before the formal analysis of structural validity, KMO and bartlett tests should be conducted on the questionnaire, and then whether it is suitable for factor analysis should be decided. KMO value is used to judge the acceptability of selected variables in factor analysis, and to investigate the correlation between variables.

Generally speaking, the kmo value of factor analysis is greater than 0.6. Besides processing, we also need to pay attention to bartlett test. In principle, bartlett test is to test whether each variable is independent and determine the correlation of factors. If the model is significant (the corresponding p value is less than 0.05), it is suitable for factor analysis.