Spearman's rho correlation is designed for use with two ordinal (a.k.a. ranked) variables. An ordinal variable is one where the values have order but you cannot do arithmetic operations with them. An example is Likert scale data. The Pearson correlation is designed for use with continuous variables (or interval or ratio data). Height and weight are examples of ratio data. Temperature in Celsius is an example of interval data.

@ Gordon L Warren ,

Thank you sir for your response but, it cannot solve my problem. Here is a paper in the attached file. The author has calculated both the correlation coefficients but he has explained the results by the most robust Spearman rank correlation coefficient. You can see line 6, paragraph 1 and line 8, paragraph,4 on page 542.

My question is 'if some one explain their results by Spearman rank correlation' then how to explain the reason of calculating both Pearson linear and Spearman rank correlation coefficient?

Thanks

Dear Abdur Rauf

There are many assumption must available to use Pearson correlation coefficient, such as the two variables must be scale variables, and normally distributed. In other hand, Spearman rank correlation used for ordinal variables.

Good Luck

@ Khalid Hassan

Sir, i appreciate your response. When you find some free time then i wish you read the few sentences highlighted in the attached file. Here in this paper the author has calculated both Pearson linear and Spearman rank correlation for same kind of parameters. He is explaining the results by using the more robust Spearman rank correlation coefficient. My question here is

" if he is explaining the results by Spearman rank correlation coefficient then why it is needed to calculate Pearson correlation?"

Thanks

Dear Abdu Rauf

Simply what are you measure? If you are interested in calculation the relationship between the values of the two variables , use the simple correlation. If you are interested in measuring the relationship between the rank of the values of the two correlation use the Spearman rank correlation.

Good luck

@Ahmed Shemeis

Sir, I have understood both the correlation up to some extent. I am studying the attached paper. I n this paper, the has calculated both the correlation and explained the results by Spearman correlation. So, the question for me is that why he is needs to calculate both.?

good one

The Pearson correlation coefficient is the most widely used. It measures the strength of the linear relationship between normally distributed variables. When the variables are not normally distributed or the relationship between the variables is not linear, it may be more appropriate to use the Spearman rank correlation method. There is a very interesting paper about the differences between these two correlation coefficients on the same sets of data:

http://geoinfo.amu.edu.pl/qg/archives/2011/QG302_087-093.pdf

Hope it will be useful!

@ Abdulvahed Khaledi Darvishan ,

Thank you sir.

The basic assumption of Pearson correlation coefficient is that it assumes constant variance and can quantify the linear association between two variables. If any deviation happens to this assumption it is worth trying the Spearman rank correlation. Although for most of the cases they have equal values, it is better to try both the methods for same sets of data.