I am defending my dissertation today and this is by far the best video I have watched that explains both tests in a way that I understand it and can talk about it in the future. Thank you!!!
Amazing description on this topic. I am working on a project and was having some difficulty in wrapping my head around the topic, but this has cleared that up. Thank you!!!
hi! im having a hard time understanding you explanation for the monotonic relation ship. the data I have looks at atmospheric pressure, in which it has a pattern of my numbers going up and down during the seasons of a year. my dependent variable are dust particles which also increase and decrease (although not as uniform as atmospheric pressure does in a year). I want to see if atmospheric pressure and the level of dust particles in the air are correlated. do I use spear sons or Pearsons?
If the correlation between the variables go up and down, with your example I would use Pearson's. That will give you more information but it also depends on what type of relationship you are trying to gauge between variables. Both Pearson and Spearman have their uses, one is more sensitive than the other but it also depends on the nature of the data.
Hi! What is the best way to plot a spearman correlation so that we clearly see the information that it is giving? I think it's not useful to plot a regression line as it depends on outliers, while the spearman correlation doesn't.
There really isn't one as an "ideal" Spearman can take many forms whereas an "ideal" Pearson's correlation is a straight line that is positive or negative 1. Regression lines are still used as they give a comparison within the graph against the Spearman correlation.
Hello, it was very helpful, I have a question about the (Zero Coefficient) in the last step. why there is no relation? if there is a nonlinear relation between variables???. Kind Regards
No problem. The Pearson Correlation equation is like a standardized slope for a linear regression line. It cares about the cov(x,y) and normality of data, so it ends up equaling 0 when the data goes down and up or up and down like that. Spearman Correlation is a more general approach because it ranks the variables and doesn't assume the normality of the data. It cares that the variables are monotonic, that is a one to one type of relationship. The quadratic creates a type of one to two because as the first variable increases the second variable goes up and down or down and up. If you find yourself needing a correlation coefficient for nonlinear models or data, Kendall Correlation would help with that. It is another nonparametric test.
Thanks a lot for your lovely explanation, my question if we plot spearman correlation can we use rs value to find r2 or it is for pearson only? Another word, is spearman can be predicted the value of variables using the r2 or slope equation ? Thanks in advance
You will not be able to convert a spearman rs to a pearson r. There may be similarities at times such as a 45 degree angle where both are 1 but what they are measuring is different and non-transferable.
I am defending my dissertation today and this is by far the best video I have watched that explains both tests in a way that I understand it and can talk about it in the future. Thank you!!!
Thank you, I hope your defense goes well.
I was having a little doubt around this topic. But, now I got that. Thanks.
Amazing description on this topic. I am working on a project and was having some difficulty in wrapping my head around the topic, but this has cleared that up. Thank you!!!
Phenomenal explanation, very well done!
Thank you! i have searched so many videos to explain Rank correlation. This is the best! 👍
You're welcome!
Support the channel and purchase from Amazon through this link with no cost to you: amzn.to/43PwRLN
Greate video, very well made!
Thank you!
Thank you. Great video :)
Awesome! Thanks
What's the correlation coefficient of spearman if one value is always the same? e.g. X = 3, 5, 1, 2, 8; Y = 0, 0, 0, 0, 0
Since Y stays at 0 and Spearman looks at the line through a monotonic lens, it would be 1.
hi! im having a hard time understanding you explanation for the monotonic relation ship. the data I have looks at atmospheric pressure, in which it has a pattern of my numbers going up and down during the seasons of a year. my dependent variable are dust particles which also increase and decrease (although not as uniform as atmospheric pressure does in a year). I want to see if atmospheric pressure and the level of dust particles in the air are correlated. do I use spear sons or Pearsons?
If the correlation between the variables go up and down, with your example I would use Pearson's. That will give you more information but it also depends on what type of relationship you are trying to gauge between variables. Both Pearson and Spearman have their uses, one is more sensitive than the other but it also depends on the nature of the data.
Helpful, Thank you 🙏🏼
Good video!
Thank you!
Hi! What is the best way to plot a spearman correlation so that we clearly see the information that it is giving? I think it's not useful to plot a regression line as it depends on outliers, while the spearman correlation doesn't.
There really isn't one as an "ideal" Spearman can take many forms whereas an "ideal" Pearson's correlation is a straight line that is positive or negative 1. Regression lines are still used as they give a comparison within the graph against the Spearman correlation.
Hello, it was very helpful, I have a question about the (Zero Coefficient) in the last step. why there is no relation? if there is a nonlinear relation between variables???. Kind Regards
No problem. The Pearson Correlation equation is like a standardized slope for a linear regression line. It cares about the cov(x,y) and normality of data, so it ends up equaling 0 when the data goes down and up or up and down like that.
Spearman Correlation is a more general approach because it ranks the variables and doesn't assume the normality of the data. It cares that the variables are monotonic, that is a one to one type of relationship. The quadratic creates a type of one to two because as the first variable increases the second variable goes up and down or down and up.
If you find yourself needing a correlation coefficient for nonlinear models or data, Kendall Correlation would help with that. It is another nonparametric test.
excellent
Thanks a lot for your lovely explanation, my question if we plot spearman correlation can we use rs value to find r2 or it is for pearson only? Another word, is spearman can be predicted the value of variables using the r2 or slope equation ? Thanks in advance
You will not be able to convert a spearman rs to a pearson r.
There may be similarities at times such as a 45 degree angle where both are 1 but what they are measuring is different and non-transferable.