Finding correlation between two groups in survey

Question

I have a friend who asked me for help in statistics. I study mathematics, but have never dealt with statistics in all of my years of study, so I don't want to give them the wrong information. Either way, here's what is needed.

They have a group of $n$ participants in a study. Some participants have property $A$, some have property $B$, and some have both. What they want to know is the best way to measure if there is a correlation between those with property $A$ and those with property $B$. The hypothesis is that if something is to have property $A$, they are more likely to have property $B$. What is the best way to test this?

Thank you in advance for your help.

Answer 1

You should use the Chi Squared test.

Create a $2$x$2$ table with colums (Not A, A) and rows (Not B, B)

You can see the rest at this link, the calculations should be fairly fast for a $2$ by $2$.

https://www.mathsisfun.com/data/chi-square-test.html

Answer 2

You need to find the correlation coefficient of the properties A and B. First find the covariance: E(AB)-E(A)E(B). Then find the Std. Dev. of each. Use the formula $\frac {Cov(A,B)}{\sigma_A \sigma_B}$. To get correlation coefficient. It lies between -1 and 1. Closer to 1 means better correlation. Closer to zero implies no correlation. Note: E(A) is expected value of A