How do I write a Results section for Correlation?
From PsychWiki  A Collaborative Psychology Wiki
(Difference between revisions)
Line 21:  Line 21:  
    
  ◄ Back to [[  +  ◄ Back to [[Writing a Paper]] page 
Latest revision as of 05:42, 17 November 2016
How do I write a Results section for Correlation?
 The report of a correlation should include:
 r  the strength of the relationship
 p value  the significance level. "Significance" tells you the probability that the line is due to chance. More specifically, the "Significance" represents a test of whether the line is different from a flat line (e.g. a flat line would be represented by a Pearson correlation = 0). Any pvalues less than .05 indicates that the result is not due to chance.

 Some researchers also include:

 n  the sample size
 Descriptive statistics of each variable
 R^{2}  the coefficient of determination. This is the amount of variance explained by another variable. You can calculate R^{2} by squaring the Pearson Correlation. If you multiple this by 100, you converted the value into a percentage.  If your correlation is .352, then .352 x .352 = .124. So the variance explained is 12.4%. This also means that 87.6% of the variance is unexplained since 100 .124 = .876. In terms of percentage of variance explained, small is 1%, medium is 9%, and large is 25%.
 Examples:
 There was a positive correlation between the two variables, r = .35, p = < .001.
 There was a positive correlation between height (M = 55.39 SD = 16.33) and weight (M = 145.22 SD = 15.54) , r = .35, p = < .001, n = 100.
 There was a positive correlation between the two variables, r = .35, p = < .001, with a R^{2} = .124
◄ Back to Writing a Paper page