五级量表法信度与效度剖析(国外英文资料).docVIP

五级量表法信度与效度分析(国外英文资料) Reliability analysis Reliability is represented by the consistency and stability of the scale, the reliability was divided into external and internal reliability reliability, reliability of the internal analysis method has a lot of, often with Cronbach alpha coefficient to estimate, the Cronbach alpha coefficient, the greater the said the greater the correlation of the variables of each item, namely, the higher the degree of internal consistency. American statistician Hair and others believe that Cronbach alpha is greater than 0.7 for high reliability. Normally, the Cronbach alpha coefficient is between 0.8 and 0.9, which is considered very good, between 0.7-0.8. This study used SPSS software to analyze the scale of the questionnaire in the questionnaire, and the Cronbachs Alpha coefficient was shown in table 1. Table 1, the Cronbachs Alpha coefficient Cronbachs Alpha A number of 0.831 18 From the table above, the Cronbachs Alpha coefficient of 18 items is 0.831, greater than 0.8, indicating that the questionnaire is highly reliable. Analysis of validity (factor) The validity of the scale is usually done by the factor analysis, and the measuring scale is suitable for the analysis of the factors, and the scales are tested by the KMO and the batley ball. KMO values are typically distributed between 0 and 1, and the closer to 1, the more suitable for factor analysis. The common criterion for factor analysis is that the KMO value is above 0.6. The sig value (that is, p value) of the batley ball test statistic is less than 0.05, which is suitable for the factor analysis when it reaches a significant level. To KMO questionnaire in scale and bartley sphericity test, the results showed KMO value of 0.632, higher than 0.6, the spherical degree of Bartlett test for sig value. 000, at 0.05 significant level, scale suitable for factor analysis. Using principal component analysis (pca) was carried out on the scale factor analysis, and through the maximum va