Correction for Restriction of Range when Both X and Y are Truncated

The effect of range restriction on one variable in a bivariate normal distribution on the X-Y correlation and the problem of estimating unrestricted from re stricted correlations has been widely studied for more than half a century. The behavior of correction formu las under truncation of both X and Y, however, re mains largely unresearched. The performance of the correction formula for unidimensional truncation (Thorndike, 1947, Case 2) and an approximation pro cedure for correcting for bidimensional truncation pro posed by Wells and Fruchter (1970) were investigated. The Thorndike correction formula undercorrects in most circumstances. The Wells and Fruchter procedure performs quite well under most conditions but often results in a slight overcorrection. The performance of the Wells and Fruchter and Thorndike formulas are also compared under truncation on X or Y alone. In these circumstances the Wells and Fruchter correction is either equal or markedly superior to the traditional correction. Based on overall performance in recaptur ing the unbiased population values under both unidi mensional and bidimensional truncation, the Wells and Fruchter correction is recommended as the preferred procedure in many practical settings.

[1]  M. R. Novick,et al.  Statistical Theories of Mental Test Scores. , 1971 .

[2]  H. G. Osburn,et al.  An Empirical Study of the Accuracy of Corrections for Restriction in Range Due to Explicit Selection , 1979 .

[3]  K. Subrahmaniam,et al.  On the sample correlation coefficient in the truncated bivariate normal population , 1978 .

[4]  M. A. Hamdan,et al.  Correlation in a Bivariate Normal Distribution with Truncation in Both Variables , 1971 .

[5]  G. V. Barrett,et al.  Further consideration of the power to detect nonzero validity coefficients under range restriction. , 1985 .

[6]  P. Walmsley,et al.  Statistical Method , 1923, Nature.

[7]  B. Fruchter,et al.  Correcting the Correlation Coefficient for Explicit Restriction on Both Variables , 1970 .

[8]  A. Gross.  Relaxing the Assumptions Underlying Corrections for Restriction of Range , 1982 .

[9]  R. L. Thorndike.  RESEARCH PROBLEMS AND TECHNIQUES , 1947 .

[10]  Interval Estimation of Correlation Coefficients Corrected for Restriction of Range , 1976 .

[11]  A. Gross,et al.  Restriction of Range Corrections When Both Distribution and Selection Assumptions Are Violated , 1983 .

[12]  Karl Pearson,et al.  Mathematical Contributions to the Theory of Evolution. XI. On the Influence of Natural Selection on the Variability and Correlation of Organs , 1903 .

[13]  G. W. Snedecor.  Statistical Methods , 1964 .

[14]  Robert L. Linn,et al.  Range restriction problems in the use of self-selected groups for test validation. , 1968 .

[15]  James S. Phillips,et al.  Concurrent and predictive validity designs: A critical reanalysis. , 1981 .

[16]  Z. W. Birnbaum,et al.  ON THE EFFECT OF TRUNCATION IN SOME OR ALL COORDINATES OF A MULTINORMAL POPULATION , 1951 .

[17]  K. Pearson.  VI. On the Influence of Double Selection on the Variation and Correlation of two Character , 1908 .

[18]  A. S. Otis.  A Method of Inferring the Change in a Coefficient of Correlation Resulting from a Change in the Heterogeneity of the Group. , 1922 .

[19]  H. Weiller.  MEANS AND STANDARD DEVIATIONS OF A TRUNCATED NORMAL BIVARIATE DISTRIBUTION , 1959 .

[20]  G. V. Barrett,et al.  Towards a general model of non-random sampling and the impact on population correlation: generalizations of Berkson's Fallacy and restriction of range. , 1986, The British journal of mathematical and statistical psychology.

[21]  G. M. Tallis.  The Moment Generating Function of the Truncated Multi‐Normal Distribution , 1961 .