Hi,
I have written about correlation in my previous post Correlation and Pearson’s correlation coefficient. I also worked out a small data set to identify the correlation coefficient. In this post, let’s see how quickly we shall generate the same using LibreOffice Calc (FOSS Equivalent of Microsoft Excel).
Here is the data set. We have two variables x and y. We have 7 samples for each.

Select Data>Statistics>Correlation

Choose all data set we have (values of x and y given above)

Click the Results to: button to decide where LibreOffice to write the output. I have chosen the cell D2.

Finally, following is my inputs. Click Ok.

Here is the output.

Column2 and column 1 is 0.95.
What does this indicate? Check the below given picture. Out correlation coefficient is 0.95 in the scale of -1 to +1. So x and y are highly positively correlated.

If this is true, the graphical representation should show a raising line, shouldn’t it? Let’s try.

Obviously!
Pingback: Linear Regression | JavaShine
Pingback: Testing of Independence – Chi Square test – Manual, LibreOffice, R | JavaShine
Pingback: Regression testing in R | JavaShine
Thanks for sharing – I was going to turn to Octave for my analysis, but your blog helped me figure out how to use LibreOffice instead. 🙂