THE-SAS-MOM
Follow Us:
  • Welcome
  • About SAS-Mom
  • SAS Blog
  • Typical SAS Interview Questions
  • SAS Certifications
  • Python Blog
  • Contact SAS-Mom

Using PROC CORR for Correlation Analysis of Continuous Variables.

2/12/2014

0 Comments

 
PROC CORR is used for performing Correlation Analysis and originates from the word Correlation. It focuses on the measure of the strength of the linear relationship between two continuous variables. Let’s recall that continuous variables are a subset of quantitative variables and that they store numerical data that is measured on a scale that has an infinite number of values containing no breaks such as 234, 234.6 and 235.0.
Picture
The relationship or association between two continuous variables can be parabolic, curvilinear, quadratic, cyclical or random among others but PROC CORR is primarily focused on the linear relationship between two continuous variables.

Picture
According to the SAS Product Documentation, PROC CORR computes the following information:

  • Tables of variable information

  • Simple descriptive statistics for each variable

  • Scatter plots

  • Scatter plot matrices 

  • Pearson Correlation Coefficients

    • Pearson product-moment correlation

       

  • 3 nonparametric measures of association

    • Spearman rank-order correlation

    • Kendall’s tau-b coefficient

    • Hoeffing’s measure of dependence, D)

       

  • Pearson, Spearman and Kendall partial correlation

     

Understanding correlation statistics for data is an integral aspect of correlation analysis. These correlation statistics measure the degree of the linear association between two variables by providing us with a numeric value that represents the strength of the association or relationship. The values of the correlation coefficient are always between -1 and +1 inclusive. A correlation coefficient of +1 indicates a positive linear association, a correlation coefficient of -1 indicates a negative linear association, and a correlation coefficient of 0 indicates no linear association between the two variables.

It is important to stress that correlation analysis of two variables does not define a cause-and-effect relationship. That is, the existence of a linear relationship between two variables does not mean that one variable has an effect on the variable. The primary concern of correlation is the measurement of the degree of the linear association.

Now, let’s get our hands dirty and dive right into PROC CORR and how it can be used to demonstrate the linear association or relationship between two continuous variables. The following SAS code obtains correlation statistics for the variable PctBodyFat1 and each of the variables Age, Weight and Height.

Picture

Picture
Picture
0 Comments



Leave a Reply.

    Author

    My name is Vanessa Afolabi also known as @TheSASMom. I am a Data Scientist fluent in SAS, R, Python and SQL with a passion for Machine Learning and Research.

    Archives

    August 2019
    January 2015
    December 2014

    Categories

    All

    RSS Feed

Powered by Create your own unique website with customizable templates.