Correlation analysis is used to study the relationship between quantitative data, including whether there is a relationship and how close the relationship is.
1. If there is significance (there is an * in the upper right corner of the result, it means there is a relationship; otherwise, there is no relationship); after there is a relationship, the closeness of the relationship can be directly determined by the size of the correlation coefficient. Generally, a value above 0.7 indicates a very close relationship; a value between 0.4 and 0.7 indicates a close relationship; a value between 0.2 and 0.4 indicates a general relationship.
2. If the correlation coefficient value is less than 0.2, but it still shows significance (there is an * in the upper right corner, one * is significant at the 0.05 level, and two *s are significant at the 0.01 level; significance refers to the occurrence of the correlation coefficient Statistically significant and common, rather than occurring by chance), indicating that the relationship is weak, but still relevant.
3. Correlation analysis is a prerequisite for regression analysis. First, it is necessary to ensure that there is a correlation, and then the regression influence relationship can be studied.
4. Because if they all show no correlation, it is impossible to have an impact. If there is a correlation, there may not necessarily be a regression relationship.
Related analysis steps
1. SPSSAUUsers can freely drag and drop analysis items into the analysis list box. The only difference is the output format.
2. Correlation analysis uses correlation coefficients to express the relationship between analysis items; first determine whether there is a relationship (an * sign means there is a relationship, otherwise it means there is no relationship);
3. Then determine whether the relationship is positive or negative (the correlation coefficient is greater than 0, which means positive correlation, and vice versa);
4. Finally, judge the closeness of the relationship (usually if the correlation coefficient is greater than 0.4, it means the relationship is close);
5. There are two common types of correlation coefficients, namely Pearson and Spearman. This system uses the Pearson correlation coefficient by default. Before correlation analysis, SPSSAU recommends using scatter plots to visually view the relationship between data. In addition, SPSSAU also provides the Kendall correlation coefficient. The difference between the three correlation coefficients is as follows:
case analysis
1. Background
For example, if you want to study the relationship between "Taobao customer service attitude", "Taobao merchant service quality" and "Taobao merchant satisfaction" and "Taobao loyalty" respectively, it can be clearly seen in this sentence "Taobao customer service attitude" and "Taobao merchant service quality" are X; while "Taobao merchant satisfaction" and "Taobao loyalty" These two terms areY.
2. Operation
X and Y are distinguished here, so just put them in accordingly. If you do not distinguish between X or Y, just put all the items into the "Analysis Item Y (Quantitative)" box at this time.
Image source: SPSSAU official help manual
3. SPSSAU output results
Image source: SPSSAU analysis results page
Image source: SPSSAU analysis results page
4. Text analysis
The above table uses correlation analysis to study the correlation between "Taobao merchant satisfaction", "Taobao loyalty" and "Taobao customer service attitude" and "Taobao merchant service quality" respectively, and uses the Pearson correlation coefficient to express the correlation. Condition. As can be seen from the above table:
"Taobao merchant satisfaction" is significantly significant (P <0.01) with "Taobao customer service attitude" and "Taobao merchant service quality", and the correlation coefficient values are all higher than 0.7, indicating that "Taobao merchants are satisfied Degree" has a very close positive correlation with "Taobao customer service attitude" and "Taobao merchant service quality" respectively. Similarly, "Taobao loyalty" will have a very close positive correlation with "Taobao customer service attitude" and "Taobao merchant service quality" respectively, with the correlation coefficient values being 0.673 and 0.606 respectively.
5. Analysis
Correlation analysis only studies whether there is a relationship or not. If there should be a relationship from common sense, then the correlation coefficient will always show significance. Generally speaking, after correlation analysis, it is necessary to study the influence relationship and use regression analysis method.
Supplementary information reference
SPSSAU-SPSS related analysis help manual
SPSSAU-SPSS Regression Analysis Help Manual
SPSSAU-correlation analysis/regression analysis difference connection
Author: SPSSAU
Link: https://www.zhihu.com/question/22114982/answer/583955025
Source: Zhihu< /span>
Copyright belongs to the author. For commercial reprinting, please contact the author for authorization. For non-commercial reprinting, please indicate the source.