Creating a correlation matrix in SPSS is a straightforward process that allows you to assess the relationships between multiple variables in your dataset. By following these steps, you can effectively create and interpret a correlation matrix in SPSS, providing valuable insights into the relationships among the variables in your dataset. Please use the SPSS data set (i.e., Environmental Responsible-390 [2024]) only after completing a factor analysis and reliability test. We will use the remaining questionnaire items from the formal factor analysis and reliability test to produce the correlation matrix results (remind: We will no longer use any deleted items from the factor analysis and reliability test for further data analyses).

I-Purposes:

A correlation matrix is a statistical tool that helps in data analysis and research by assessing the relationships between variables. It helps identify patterns and evaluate the strength of these relationships, aiding in variable selection for regression analysis. It also detects multicollinearity among independent variables, impacting results reliability. It facilitates hypothesis generation, exploring potential causal relationships. The matrix enhances data visualization through heatmaps, making complex data sets more interpretable. It also guides variable inclusion in predictive models based on their relationships with the dependent variable. Overall, it supports robust data-driven conclusions. The correlation coefficient, named after Karl Pearson (1900), measures the strength of the relationship between two sets of interval-scaled or ratio-scaled variables, ranging from -1.00 to +1.00 inclusive, with a value of -1.00 or +1.00 indicating perfect correlation (Lind et al., 2019). The simple correlation coefficient between two variables is a measure of the linear relationship between the variables and is always between −1 and 1. This implies that the square of the simple correlation coefficient is between 0 and 1. The nearer that the square of the simple correlation coefficient is to 1, the stronger is the linear relationship between the two variables (Bowerman et al., 2019).

II-Rule of Thumbs: Correlation Matrix

In statistics, the correlation coefficient quantifies the degree to which two variables are related. The range of the Pearson correlation coefficient (r) is from -1 to 1. Here’s a general guide for interpreting the strength of correlations and other common ranges include (Table 1). Threshold values can vary depending on the context and the specific field of study, but the ranges above are widely accepted in many applications.

Table 1. General Rule of Thumbs

Correlation	Very Weak	Weak	Moderate	Strong	Very Strong
Negative (-)	-0.1<	-0.3 < r ≤ -0.1	-0.5 < r ≤ -0.3	-0.7 < r ≤ -0.5	-1 ≤ r ≤ -0.7
Positive (+)	r < 0.1	0.1 < r ≤ 0.3	0.3 < r ≤ 0.5	0.5 < r ≤ 0.7	0.7 ≤ r ≤ 1

1. Compute Mean Score

We need to compute the mean score for individual research constructs and sub-dimensions based on the formal results of factor analysis and reliability tests. The purpose of computing the mean score is to produce the results of the correlation matrix and regression analysis. For step-by-step instructions on this stage, refer to the Appendix.

Go to Transform >> Compute Variable (Figure 1.1) >> in “Target Variable” box, you can type any abbreviation for individual research constructs or factors or sub-dimension (for this tutoring “Environmental Knowledge” will replace ENK for SPSS coding. In “Numeric Expression” box, you can write down and insert questionnaire items after a formal test of factor analysis and reliability test: Mean (ENK1, ENK2, ENK3, ENK4) or you can use another style, Mean (ENK1+ ENK2+ENK3+ENK4)/4 (Figure 1.2) >> OK. Then, the results of these two computing styles will be consistent (Refer to Figure 1.3). Finally click on “Data View” to see the results of computing mean score.

2. Correlation Matrix

After computing the mean score of individual research constructs and subdimensions, We can then proceed to generate the correlation matrix result. For step-by-step instructions on this stage, refer to the Appendix.

Go to Analyze >> Correlate >> Bivariate (Figure 2.1) >> move the mean scores of individual research constructs or sub-dimension to “Variable” box >> Options >> check on “Means and Standard Deviations” >> Continue >> OK. Then, you will see the outputs of correlation matrix (Figure 2.3).

3. Result and Interpretation

Correlation matrix is the standard form for reporting correlation coefficients for more than two variables (Zikmund et al., 2013). Correlation is the measure of the size and direction of the linear relationship between the two variables, and squared correlation is the measure of strength of association between them (Tabachnick et al., 2018). The results of Table 2 indicates that all research variables of this study have a positive significant relationship with other research variables at significant level of ^**p-value < 0.01 with Pearson Correlation coefficient test of two-tailed test. In other words, the correlation matrix was used to evaluate the correlation between the variables (Steiger, 1980). Correlation matrix illustrates the inter-relationship among key research variables as proposed in the conceptual framework (Ngounhort et al., 2024). When writing the results of a correlation matrix in APA style, you should report the correlation coefficients and their significance clearly and concisely. A Pearson correlation analysis (Table 2) was conducted to assess the relationships between variables (i.e., Environmental Knowledge, Environmental Awareness, Pro-Environmental Attitude, and Environmental Responsibility). The results indicated that there was a statistically significant positive correlation between Environmental Knowledge and Environmental Awareness (r = 0.848^**, p < 0.01), suggesting that as Environmental Knowledge increases, Environmental Awareness also tends to increase. Additionally, a moderate positive correlation was found between Environmental Knowledge and Pro-Environmental Attitude (r = 0.669^**, p < 0.01), indicating that higher values of Environmental Knowledge are associated with higherr values of Pro-Environmental Attitude. Interestingly, Environmental Awareness has strongly positive correlation with Pro-Environmental Attitude (r = 0.691^**p < 0.01), suggesting that as Environmental Awareness increases, Pro-Environmental Attitude also tends to increase. In conclusion, the findings suggest that the local community’s environmental awareness significantly influences their perception of pro-environmental attitudes, accounting for 69.1% of the variance observed in this relationship. This substantial percentage indicates that enhancing environmental awareness within the community is imperative for fostering a stronger commitment to pro-environmental behaviors and attitudes. Therefore, efforts aimed at increasing environmental awareness can be an effective strategy for promoting sustainable practices and encouraging a collective responsibility towards environmental stewardship among community members.

Table 2. The Result of Correlation Matrix (n=390)

Variables	Mean	Std.D	ENK	ENA	PEA	GPB	GRB	COM
ENK	3.51	1.11	1.00
ENA	3.52	1.04	0.848**	1.00
PEA	3.26	0.98	0.669**	0.691**	1.00
GPB	3.49	0.91	0.721**	0.714**	0.570**	1.00
GRB	3.51	0.98	0.867**	0.824**	0.691**	0.757**	1.00
COM	3.46	0.99	0.736**	0.748**	0.688**	0.767**	0.88**	1.00

4. Appendix: Step by Step-Correlation Matrix

5. Tutoring Session: Correlation Matrix

6. References

1-Bowerman, B. L., Hummel, R. M., Drougas, A. M., Moninger, K. B., Duckworth, W. M., Schur, P. J., & Froelich, A. G. (2019). Business statistics and analytics in practice (9th ed.). McGraw-Hill Education.
2-Lind, D. A., Marchal, W. G., & Wathen, S. A. (2019). Basic statistics for business & economics (9th ed.). McGraw-Hill.
3-Ngounhort, H., Chanveasna, U., Kirivadid, K., & Veasna, S. (2024). The Antecedents and Consequences of Job Satisfaction on Teachers’ Job Retention in HEI, Cambodia. Open Journal of Social Sciences, 12(9), 1-33.
4-Steiger, J. H. (1980). Tests for Comparing Elements of A Correlation Matrix. Psychological Bulletin, 245-251.
5-Tabachnick, B. G., Fidell, L. S., & Ullman, J. B. (2018). Using multivariate statistics (7th ed.). Pearson Boston, MA.
6-Zikmund, W. G., Babin, B. J., Carr, J. C., & Griffin, M. (2013). Business research methods. Cengage learning.

Tips and Further Materials

Step 1: Prepare Your Data

1. Open SPSS and load your dataset.
2. Ensure that the variables you want to analyze are appropriate for correlation (i.e., they should be measured at the interval or ratio level).

Step 2: Conduct Correlation Analysis

1. In SPSS, go to the menu and click on Analyze.
2. Select Correlate and then choose Bivariate.
3. In the dialog box, move the variables you want to include in the correlation matrix into the “Variables” box.
4. Under Correlation Coefficients, make sure “Pearson” is selected (you can also check “Spearman” or “Kendall” if your data are ordinal or not normally distributed).
5. Optionally, check the box for Flag significant correlations to highlight correlations with statistical significance in the output.
6. If you want to see two-tailed or one-tailed significance, you can make that choice in the same dialog.
7. Click OK.

Step 3: Interpret the Output

1. SPSS will generate an output window that displays the correlation matrix.
2. The correlation coefficients will range from -1 to 1:
– Number 1 indicates a perfect positive correlation,
– Number -1 indicates a perfect negative correlation,
– Number 0 indicates no correlation.
3. The diagonal of the matrix will contain 1s, as each variable is perfectly correlated with itself.
4. Look for the Sig. (2-tailed) column to find the p-values associated with each correlation coefficient. Typically, a p-value less than 0.05 is considered statistically significant.

Example Output Interpretation

1. Suppose you have a correlation coefficient of 0.75 between Variable A and Variable B with a p-value of 0.001. This indicates a strong positive relationship between the two variables, and the relationship is statistically significant.
2. Conversely, if you find a correlation of -0.3 between Variable C and Variable D with a p-value of 0.08, this indicates a weak negative relationship, but it is not statistically significant at the 0.05 level.