Correlation of variables in SPSS

By Priya Chetty on July 15, 2015

It measures the correlations between two or more numeric variables. There are two types of correlations; bivariate and partial correlations. While Bivariate Correlations are computed using Pearson/Spearman Correlation Coefficient wherein it gives the measure of correlations between variables or rank orders. The value of the correlation coefficient lies between -1 to +1 and the value “0” indicates that there is no correlation. On the other hand, the Partial correlation procedure is applied to calculate the partial correlation coefficient in order to describe the relationship between two variables along with adjustments made regarding the effect of one variable on another.

While Pearson Correlation Coefficient is used to calculate linear relationships, i.e. if one variable increases or decreases what is the extent to which other variable also increases or decreases? For example, to determine the relationship between a rising temperature and a decrease in the level of snow we would use the Pearson correlation.

On the other hand, the Spearman Correlation coefficient is used to determine the monotonic relationship between variables. For example; to determine the time taken by employees to complete the test depending on the time they have been employed within the firm. However, between the two methods, Pearson correlation is found to be a more precise method to determine correlations.

EXAMPLE

Bivariate: Pearson Correlation coefficient is calculated to determine the relationship (weak/strong) between the current salary and the beginning salary of employees within the organization.

Partial Correlation: Partial correlation can be applied to study the correlation between current salary and beginning salary along with control variables like; months since hired, and previous experience of the employee.

Bivariate correlations

Bivariate correlations
Select Analyze, -> correlate and then “Bivariate”, which opens a dialogue box as shown.
Offer ID is invalid
Select Current Salary and Beginning Salary and move them to VARIABLE list, as shown in below (Figure 2).
Select Current Salary and Beginning Salary and move them to the VARIABLE list, as shown.
NOTE

With respect to the One-Tailed and Two-tailed test of significance, in the case of the one-tailed test, you as a researcher are proposing that there is one possible relationship between the two variables which are being tested and that either the variable is higher than the mean or lower than mean. However, in the case of a two-tailed test, we are speculating that these variables can be both higher than the average mean or lower than the average mean. We generally use the Two-Tailed Test as it has better precision.

Correlations

Current SalaryBeginning Salary
Current SalaryPearson Correlation1.573*
Sig. (2-tailed).045
N255255
Beginning SalaryPearson Correlation.573*1
Sig. (2-tailed).045
N255255
**. Correlation is significant at the 0.01 level (2-tailed). (This means the value will be considered significant if is between 0.001 to 0,010, See 2nd example below).
*. Correlation is significant at the 0.05 level (2-tailed). (This means the value will be considered significant if is between 0.010 to 0,050).

The output table shown above provides Pearson Correlations between the pair i.e. Current Salary and Beginning Salary. The results indicate that the Beginning salary influences the current salary of the employee (α= .573, p=0.45).

NOTE

The value of Pearson correlation and Sig are complementary to each other, further and therefore both values are considered important when writing results.

Where the value of Significance (2-Tailed) is less than 0.05 significant at a 95% confidence interval. The correlation value (α) and Sig value (p) will always complement each other, i.e. if one value is “acceptable” then the other will be “acceptable” too.

Get a comprehensive literature survey of 15001 - 20000 words for 65000.00 ready in 50 days as standard order. Use 5E25A5EE63214 for a discount of 5000.
Order now

Partial correlations using SPSS

Calculating correlation coefficient in SPSS

Next, Select, Months since Hire and Previous Job Experience and Move them to the control FOR list. Click OK to Run the Procedure.

Correlations
Control VariablesCurrent SalaryBeginning Salary1.000.711**
Months Since Hired & Previous Job ExperienceCurrent SalaryCorrelation
Significance (2-tailed).001
df0251
Beginning SalaryCorrelation.711**1.000
Significance (2-tailed).001
df2510
**. Correlation is significant at the 0.01 level (2-tailed).
*. Correlation is significant at the 0.05 level (2-tailed).

Legends:

  • Pearson Correlation: Gives the value for Correlation at a confidence interval of 95%
  • Sig (2-tailed): Gives the value of significance of the correlation between the two variables at a 95% confidence interval
  • Df: Displays the degree of freedom, i.e. the sample size of the study.

The Output given above shows the table of the Partial Correlation Coefficient, the Degree of Freedom, and the significance level for the two correlating variables i.e. current Salary and Beginning salary. The results of partial correlation indicate that control variables Months since hired and Previous job experience influence the current and the beginning salary of the employees. Here α=.711 and p= 0.001 (p < 0.01) which reflects significance at 98% confidence interval.

NOTES

Discuss

11 thoughts on “Correlation of variables in SPSS”