1

I'm confused about finding correlation in these two cases with using Excel: 1- Correlation of two binary variables : (Male/Female,Interest to Coffee(Yes/NO)) 2- Correlation between one continuous and one binary : (Age,Interest to Coffee (Yes/NO))

Would you tell me or show me clear example since i'm beginner in statistics. I use Excel.

Dawny33
  • 2,239
  • 1
  • 21
  • 37
Raffael Edu
  • 21
  • 1
  • 5
  • I think you can use this formula to calculate by hand $r = r_{xy} =\frac{\sum ^n _{i=1}(x_i - \bar{x})(y_i - \bar{y})}{\sqrt{\sum ^n _{i=1}(x_i - \bar{x})^2} \sqrt{\sum ^n _{i=1}(y_i - \bar{y})^2}}$ Also read https://en.wikipedia.org/wiki/Pearson_product-moment_correlation_coefficient – Deep North Aug 12 '15 at 04:13
  • Note that for binary variables the correlation bounds can be ver different from +1 or -1. –  Feb 04 '16 at 07:10

1 Answers1

2

Correlation of two binary variables You might want to look at the phi coefficient. It is a measure of association between two binary variables. The formula and the details are better explained here

Correlation between one continuous and one binary You might want to check @kjetil b halvorsen's answer here, about the fallacies of using Spearman coefficient and also discusses a nice alternative approach.

Community
  • 1
Dawny33
  • 2,239
  • 1
  • 21
  • 37
  • thanks, but both of them are confusing for me! is there any simple example available for both calculations? I couldn't understand these procedures! – Raffael Edu Aug 12 '15 at 06:08
  • Okay, I understand. It would be better if you look at the Pearson Correlation Coefficient (https://en.wikipedia.org/wiki/Pearson_product-moment_correlation_coefficient) This is a very common method for calculating association between two variables. This (https://support.office.com/en-za/article/PEARSON-function-0c3e30fc-e5af-49c4-808a-3ef66e034c18) might help you in implementing it in Excel. And, for understanding the Spearman measure, you can refer to (http://www.real-statistics.com/correlation/spearmans-rank-correlation/). After understanding it, you would understand kjetil's answer too – Dawny33 Aug 12 '15 at 06:14