# Bivariate Correlation Spss

I need help with SPSS software and a small element dealing with writing a paper with the results. I am an undergrad.

I was asked to “Include the three main variables for the correlations: PSS, Intensity of Experience Scale, and Health” ( I can provide the files if needed)

Do I do a bivariate correlation and add all 3 of the variables to the right of the boxes to create an output?

I need a bit of help reading the results as well. I wondering if I need to identify specifically:

PSS to IEScale

Health to IEScale

PSS to Health

Am I missing anything else from the variables that correlate?

Also, in the APA paper example the professor provided, the way the output was written was:

r(84)=-.34,p=.01 (these numbers are associated with an entirely different set of data)

I am wondering if the 84 is N, the -.34 is the pearson correlations and the .01 is the significant. Is that how they set it up? If not, how do I set up my results to mimic this particular layout provided as an example we must follow.

Solution

I will answer the questions in the order they were asked:

1. Correlations. There are two types of correlation – bivariate and partial. Most probably you are requested to run the bivariate correlations between the variables. The fact that the name is BIvariate should not stop you to place more than 2 variables in the SPSS box. It simply means that the correlation coefficients that you will obtain gives the correlation between each two pairs. So yes, place the three variables in the box for Bivariate Correlation in SPSS. You will get a 3×3 matrix where in the diagonal you will have 1, the upper and the lower part of the matrix (below and above that diagonal) are equal. So using the lower semi matrix you will have 3 coefficients corresponding to the 3 correlations between the each pair comprised by those three variables.
2. Yes, in writing the results you should specify the coefficients corresponding to the specific pair, as you listed them. As you probably know the correlation coefficient can vary bteween -1 which is the perfect negative correlation to +1 which is the perfect positive correlation. The matrix also gives you the significance of each coefficient (just below the value of the coefficient) denoted in the results by p-value.
3. Almost correct. -.34 is the correlation coefficient value, .01 is the significance of the coefficient, which in this case is significant at the 5% level since .01 is < than .05. And 84 is the degrees of freedom, which equals N-2, so in this case the sample size is 86.