关于SPSS(国外英文资料).docVIP

关于SPSS(国外英文资料) The reliability analysis: Reliability refers to the use the same measurement tools to repeated measurements on the same object, the conclusions and the results of the consistency of the data, namely the stability of measurement results (stability), consistency, consistency and reliability evaluation (reliability). Generally speaking, the reliability coefficient is not directly calculated, and we can only estimate it in an indirect way. We compared the commonly used evaluation index is used to evaluate the consistency of the Cronbachs alpha values, the alpha value can more accurately reflect the measurement item of consistency, but also can reflect the internal structural. The greater the alpha, the greater the correlation between the measured indicators and the greater the reliability. In general, the alpha value should be greater than 0.70 (Nummally, 1978) in the actual application. In general exploratory research, the Cranbachs a coefficient was above 0.6 and the baseline study was above 0.8, and the Cranbachs a coefficient was more than 0.6 and was considered to be more reliable. This is based on the concept of Nunnally (1978). At the same time, we can take advantage of each item and the overall correlation coefficient (CITC) if we want to purify the item. CITC (correction of total correlation), in practice, should be greater than 0.5, at least more than 0.35 (here should indicate the source), if less than 0.35, then consider deleting this item. In other words, CITC should be deleted to a certain extent if it is small and picked out, which will cause the Cronbachs alpha value to be added to a certain degree. Validity analysis: Validity is an estimate of the effectiveness of the measurement results, which is the core measure of measuring tool. If not, all indicators, including reliability, will become meaningless (Dieterish freeman, 1979). Generally speaking, the higher the validity, the more the measurement results will reflect the actual situatio