Validity, also known as validity, refers to the degree to which a measuring tool or means can accurately measure what needs to be measured. Validity refers to the degree to which the measured results reflect the intended content to be examined. The more consistent the measured results are with the content to be examined, the higher the validity; On the contrary, the lower the validity.

    Exploring factor analysis is to measure the structural validity of the scale and to determine whether the measurement variables of each latent variable have stable consistency and structure. It is the most commonly used indicator for evaluating the validity of the scale. This article tests the composition of each dimension. When using factor analysis for validity analysis, the first step is to determine whether the conditions for factor analysis are met. Generally, two conditions need to be met, one of which is to require a KMO value greater than 0.7; The second reason is that the significance of Bartlett's sphericity test is less than 0.05. If these two conditions are met, it indicates a strong correlation between the observed variables and is suitable for factor analysis.

    Data description:

    Here, A1~A10 are used for analysis, and the analysis table is as follows:

    From the validity analysis, it can be seen that the KMO and Bartlett tests were used for validity verification. From the table above, it can be seen that the KMO test value of the survey data is 0.805, which is greater than 0.7, indicating that the questionnaire is suitable for factor analysis. The results of Bartlett's sphericity test show that the approximate chi square value is 287.201, and the probability of significance is 0.0. Therefore, the null hypothesis of Bartlett's sphericity test is rejected, and the scale is considered suitable for factor analysis. Therefore, the validity structure is good.

    The following is an explanation of the calculation result indicators:

    The Bartlett's spherical test method is based on the correlation coefficient matrix. Its null hypothesis is that the correlation coefficient matrix is a unit matrix, that is, all elements on the diagonal of the correlation coefficient matrix are 1, and all elements on the non diagonal are zero. The statistics of the Bartlett's spherical test method are obtained based on the determinant of the correlation coefficient matrix. If the value is large and its corresponding probability value is less than the specified significance level, the null hypothesis is rejected, Indicates that the correlation coefficient matrix is not a unit matrix and there is correlation between the original variables, making it suitable for principal component analysis; On the contrary, the null hypothesis holds, and there is no correlation between the original variables, making the data unsuitable for principal component analysis.

    The approximate chi square value is a statistical measure of Bartlett's sphericity test, which represents the degree of difference between the covariance matrix and the identity matrix of the observed data. The calculation of approximate chi square values is based on the size of the sample and the eigenvalues of the covariance matrix. If the approximate chi square value is large, it means that there is a significant difference between the covariance matrix and the identity matrix of the observed data, rejecting the original hypothesis.

    Df (degree of freedom) is the degree of freedom for Bartlett's sphericity test, which represents the degree of freedom of the covariance matrix. The calculation of df is based on the number of variables. The greater the degree of freedom, the larger the sample size is enough to conduct more accurate testing.

    The P-value is the significance level of Bartlett's sphericity test, which represents the probability of current differences in observed data under the assumption of multivariate normality. If the P-value is less than the significance level (usually 0.05), the hypothesis of multivariate normality is rejected, indicating that the observed data does not meet the hypothesis of multivariate normality.

    KMO value (Kaiser Meyer Olkin value) is an indicator used to evaluate the applicability of variables and sample suitability in factor analysis or structural equation models. The range of KMO values is between 0 and 1, and the closer it is to 1, the higher the correlation between variables, making it suitable for factor analysis or structural equation modeling.

    The calculation of KMO values is based on the correlation matrix between variables. It measures the commonality between variables, that is, they collectively explain the proportion of total variance. The higher the KMO value, the stronger the correlation between variables, making it suitable for factor analysis or structural equation modeling.

关注微信公众号发送【示例数据】获取SPSSMAX练习示例数据。