{"title":"Choosing between the Regression Correlation, the Rank Correlation, and the Correlation Curve","authors":"Roger L Goodwin","country":null,"institution":"","volume":101,"journal":"International Journal of Humanities and Social Sciences","pagesStart":1699,"pagesEnd":1704,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/10001557","abstract":"This paper presents a rank correlation curve. The\r\ntraditional correlation coefficient is valid for both continuous\r\nvariables and for integer variables using rank statistics. Since\r\nthe correlation coefficient has already been established in rank\r\nstatistics by Spearman, such a calculation can be extended to\r\nthe correlation curve.\r\nThis paper presents two survey questions. The survey\r\ncollected non-continuous variables. We will show weak to\r\nmoderate correlation. Obviously, one question has a negative\r\neffect on the other. A review of the qualitative literature\r\ncan answer which question and why. The rank correlation\r\ncurve shows which collection of responses has a positive\r\nslope and which collection of responses has a negative slope.\r\nSuch information is unavailable from the flat, \u201dfirst-glance\u201d\r\ncorrelation statistics.","references":null,"publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 101, 2015"}