Skip to main content

首页 / Posts

【SPSS】Correlation Analysis

This article introduces how to use SPSS for correlation analysis and how to interpret correlation matrices and heatmaps, helping readers quickly master correlation-analysis methods and result interpretation among scale dimensions.

Rosetears·
··499 words·3 mins

Steps
#

  1. Calculate variables:

    1. Transform > Compute Variable
    2. Enter the dimension name as the target variable, and enter mean(first item under the dimension to last item under the dimension) as the numeric expression.
      • For example, you do not need to type the item names manually; just click the corresponding labels on the left:
        image|450
    3. Perform the same operation for all dimensions.
  2. Choose the analysis method: click “Analyze” > “Correlate” > “Bivariate” in the menu bar.

  3. Select variables: in the dialog box, add the variables whose correlations you want to analyze, usually all scale dimensions, to the “Variables” box.

  4. Choose correlation coefficient type: in the “Correlation Coefficients” section, usually select Pearson’s correlation coefficient. If the data do not meet the normal-distribution assumption, choose Spearman or Kendall.

  5. Run the analysis: after setting everything, click “OK” to run the analysis.

  6. View results: SPSS generates a correlation coefficient matrix in the output window, including the correlation coefficient and significance level (p-value) for each pair of variables. Usually, p < 0.05 indicates a significant correlation.

  7. Table: draw the table in the following style:

    image.png

  8. Heatmap: you can also draw a heatmap to make the result more intuitive and visually pleasing:

    Correlation analysis heatmap.png

Plotting code

Interpretation
#

*: p < 0.05 **: p < 0.01 ***: p < 0.001

Word Explanation
#

CorrelationPerformance expectancyEffort expectancyFacilitating conditionsSocial influencePerceived riskHedonic motivationPrice valuePersonal innovativenessBehavioral intention
Performance expectancy1
Effort expectancy0.712**1
Facilitating conditions0.687**0.745**1
Social influence0.651**0.731**0.817**1
Perceived risk0.451**0.499**0.544**0.602**1
Hedonic motivation0.743**0.701**0.781**0.757**0.570**1
Price value0.646**0.618**0.766**0.691**0.593**0.752**1
Personal innovativeness0.696**0.745**0.713**0.744**0.589**0.806**0.716**1
Behavioral intention0.684**0.665**0.801**0.759**0.576**0.782**0.834**0.779**1

Correlation analysis heatmap.png
When a correlation coefficient is greater than 0, it indicates a positive correlation; when it is less than 0, it indicates a negative correlation. When its absolute value is below 0.3, there is no correlation; 0.3~0.5 indicates a low correlation; 0.5~0.8 indicates a moderate correlation; and above 0.8 indicates a high correlation. Therefore, based on the table above, the conclusions are:

The correlation coefficient between performance expectancy and effort expectancy is 0.712, P < 0.01, indicating a moderate positive correlation.

The correlation coefficient between effort expectancy and facilitating conditions is 0.745, P < 0.01, indicating a moderate positive correlation.

The correlation coefficient between facilitating conditions and social influence is 0.817, P < 0.01, indicating a high positive correlation.

The correlation coefficient between social influence and perceived risk is 0.602, P < 0.01, indicating a moderate positive correlation.

The correlation coefficient between hedonic motivation and price value is 0.752, P < 0.01, indicating a moderate positive correlation.

The correlation coefficient between personal innovativeness and behavioral intention is 0.779, P < 0.01, indicating a moderate positive correlation.

The correlation coefficient between behavioral intention and price value is 0.834, P < 0.01, indicating a high positive correlation.


Click this text to watch the video version of this article

Related

【SPSS】Difference Analysis

··1475 words·7 mins
This article explains independent-samples t tests and one-way analysis of variance (ANOVA) in difference analysis, including SPSS operation steps, homogeneity-of-variance testing, t-value analysis, and result interpretation, helping readers understand how to compare mean differences between groups in data analysis.

【SPSS】Normality Test

··268 words·2 mins
This article introduces how to use SPSS for descriptive statistical analysis, focusing on how to test normal distribution with histograms and Q-Q plots, helping users better understand data distributions and the analysis process.

【SPSS】Validity Analysis

··1081 words·6 mins
This article introduces common SPSS validity-analysis methods, including exploratory factor analysis, principal component analysis, and confirmatory factor analysis. It explains how to interpret the KMO test and factor loadings to help readers understand validity assessment in data analysis.

【SPSS】Descriptive Statistics

··3196 words·16 mins
SPSS descriptive statistics: including single-choice questions, multiple-choice questions, and crosstabs.

About Me | The Lost Dandelion

··368 words·2 mins
An introduction to the author and this small digital site, recording the starting point for learning, code, notes, and personal expression.