Laserfiche WebLink
<br />. <br /> <br />t <br /> <br /> <br />BIOMETRICS - LINEAR CORRELA nON <br /> <br />If a correlation coefficient is computed and is <br />of low magnitude, test it to determine whether <br />it is significantly different from zero. The test, a <br />t-test, is computed as follows: <br /> <br />The computed t is compared with the tabular t <br />with n- 2 degrees of freedom and chosen proba- <br />bility level. If the computed t exceeds the <br />tabular t, the null hypothesis that the true corre- <br />lation coefficient equals zero is rejected, and the <br />computed r may be used. <br /> <br />t = <br />~(l-r2) <br />(n-2) <br /> <br />(49) <br /> <br />8.0 BIBLIOGRAPHY <br /> <br />Cochran, W. G. 1959. Sampling techniques. Jolm Wiley and Sons, New York. 330 pp. <br />Li, J. C. R. 1957. Introduction to statistical inference. Edward Brothers, Inc., Ann Arbor. 553 pp. <br />Natrella, M. G. 1963. Experimental statistics. National Bureau of Standards Handbook No. 91, U.S. Govt. Printing Office. <br />Snedecor, G. W., and W. G. Cochran. 1967. Statistical methods, 6th edition. Iowa State Univ. Press, Ames. <br />Southwood, R. T. E. 1966. Ecological methods with particular reference to the study of insect populations. Chapman and Hall, Ltd., <br />London. 391 pp. <br />Steele, R. G. D., and J. H. Tonie. 1960. Principles and procedures of statistics with special reference to the biological sciences. McGraw <br />Hill, New York. 481 pp. <br />Stuart, A. 1962. Basic ideas of scientific sampling. ~fafner, New York. 99 pp. <br /> <br />27 <br />