Mr. Ali,

As the first column is not value SPSS will give you error.

If you are only concern with correlation than you can simply use excel (second link).

Please refer to some useful links below and apply proper statistics as per your requirement.

https://www.spss-tutorials.com/spss-correlation-test/

https://www.spss-tutorials.com/correlation-coefficient-what-is-it/

Hope this will help.

Ali, I do not use SPSS but R, but I hope this could help you. I created a file called 'in.txt' with the data example you give, and within R I did:

> data = read.table('in.txt', header=T)

> data

trt1 trt2 trt3 trt4

p1 1 1 0 1

p2 1 1 1 1

p3 1 1 1 1

p4 1 1 1 1

p5 1 1 1 1

p6 1 1 1 1

p7 1 1 0 1

p8 1 1 0 0

p9 1 1 1 1

> cor(data)

trt1 trt2 trt3 trt4

trt1 1 NA NA NA

trt2 NA 1 NA NA

trt3 NA NA 1.0 0.5

trt4 NA NA 0.5 1.0

Warning message:

In cor(data) : the standard deviation is zero

Columns trt1 and trt2 has no variance, so no correlation involving those columns could be calculated. However, for columns trt3 and trt4 you can see results. Of course I assume this is just an example, and that you have real data with values with more variance. By default the cor() function in R uses Pearson, but you can change that with cor(data, method="spearman") for Spearman, for example.

I hope that helps. Best regards.

Try tetrachoric correlation coefficients. psych package implemented it.

Thank you all for your swift responses.

I'll use your tips and suggestions and see how it goes.