Reliability analysis in the questionnaire survey is an indispensable indicator, representing the credibility and reliability of your questionnaire. It is most commonly used in the analysis of questionnaire scales, and is also often used in improved scoring tables. The main principle is to perform repeated and multiple measurements on an object. If the results are close every time, the reliability is considered reliable. The most commonly used one is Cronbach α reliability.
Today we are using SPSS's own product sales survey data to demonstrate Cronbach's alpha reliability analysis. First import the data as shown in the figure below: There
are a lot of data, and the previous structure is the same as the general data analysis structure. The latter are the six indicators, Price satisfaction, Variety satisfaction, Variety satisfaction, Organization satisfaction, Organization satisfaction, Service satisfaction, Item quality satisfaction, Item quality satisfaction, Overall satisfaction, and overall satisfaction
is Li Ke The classic 5-scale scale is rated as: 1 strongly negative, 2 a little negative, 3. neutral, 4. a little positive, 5. strongly affirmative, which means that consumers’ opinions are divided into 5 levels from very dissatisfied to very satisfied.
Next, let’s carry out reliability analysis.
Click Analysis------Scale------Reliability Analysis.
Select all the following 6 indicators, and do not need to change the others.
Then click Statistics, and the scale
results after selecting the item to be deleted are as follows It shows that
Kronbach is the reliability we need. Generally, it needs to be greater than 0.7 to indicate that the reliability of the questionnaire is reliable. In our case, 0.827 has far exceeded it, indicating that the reliability is very good.
If the result is less than 0.7 and does not meet the requirements, we have to analyze each item of reliability. If we delete the third item in the figure below, the total reliability will increase to 0.86, and we have to consider whether it is necessary Delete it.
For more exciting articles, please pay attention to the public number: zero-based scientific research
