Edward E. Cureton

The upper and lower twenty-seven per cent rule

A simplified re-derivation of the formula underlying the rule is presented, followed by a derivation of the comparable rule for the unitrectangular distribution, which turns out to be a 33-per cent rule. Critical comments are offered concerning two assumptions: normality of the score distribution and equality of mean standard errors of measurement in the high and low groups.

Validity, Reliability, and Baloney

as applied in certain particular situations. In the present study the criterion is college scholarship, as given by the usual gradepoint average. The subjects were 29 senior and graduate students in a course in Psychological Measurements. These students took Forms Q and R of the Cooperative Vocabulary 7est, Form R being administered about two weeks after Form Q. The correlation between grade-point average and the combined score on both forms of this test was -23. The reliability of the test, estimated by the Spearman-Brown formula from the correlation between the two forms, was .90. The experimental form of the new test, which I have termed the &dquo;B -Projective Psychokinesis Test,&dquo; or Test B, was also applied to the group. This experimental form contained 85 items, and there was a reaction to every item for every student. The items called for unequivocal &dquo;plus&dquo; or &dquo;minus&dquo; reactions, but in advance of data there is no way to tell which reaction to

Corrected item-test correlations

Previous papers on this subject derive the correlation between an item and the remainder of the test. This correlation is unsatisfactory because the reliability of the remainder varies inversely with the reliability of the item omitted. The present paper derives the correlation between an item and the total test, with that item replaced by a rationally equivalent item. The general formula is then modified, for dichotomus items, to give the corrected point-biserial, biserial, and Brogden biserial correlations. The results apply strictly only to factorially homogeneous tests: those in which the same trait or combination of traits is measured (apart from error) by every item.

Rank-biserial correlation

A formula is developed for the correlation between a ranking (possibly including ties) and a dichotomy, with limits which are always ±1. This formula is shown to be equivalent both to Kendallsτ and Spearmansρ.

An extended table of chi-square for two degrees of freedom, for use in combining probabilities from independent samples

A table of values of Chi-square for two degrees of freedom corresponding to values of P from .001 to .999 is presented, together with a description and an example of its use in combining probabilities from two or more independent samples to obtain an aggregate probability.

Reliability and validity: Basic assumptions and experimental designs.

ally independent both of one another and of the true scores. If we confine attention to the linear case, we can read &dquo;uncorrelated&dquo; for &dquo;statistically independent.&dquo; Sampling of the literature indicates that few investigators realize that these assumptions imply definite restrictions upon the experimental designs which are acceptable in research based on psychological and educational test scores when the test reliabilities are to be taken into account.

The Definition and Estimation of Test Reliability

empirical judgment, and different authors have given conflicting definitions and conflicting criteria to guide empirical judgment. Under several of these definitions, however, it can be shown that the reliability coefficient is the variance ratio of true scores to total scores on each of the equivalent forms. The present writer prefers to reverse this procedure, defining the reliability coefficient of any one form of a test as the variance ratio of true scores to total scores. This shifts the ground of a difficulty indicated by Kelly [6], that under the correlation definition we cannot discuss directly the reliability of one form of a test. Under the proposed definition such discussion is entirely possible so long as it remains formal, and we see that the second form is needed only when we come to the estimation of the

Note on φ/φ max

Formulas are given for a descriptive statistic related to the fourfold-point correlation but having always-attainablelimits ±1.

Factor Analysis of Emotionality and Conditioning in Mice

14 measures of “emotional” behavior, 3 of escape and 3 of avoidance conditioning were obtained on 126 offspring of 101 ♂ and C3H ♂ mice. A factor analysis of the data produced 4 factors, labeled “Fear I: freezing,” “Avoidance Conditioning,” “Escape Conditioning,” and “Fear II: non-locomotor variety,” respectively. The conclusions were that: (a) “emotionality” is not a unitary variable; (b) in terms of individual difference variables escape and avoidance conditioning are not related; and (c) several indices of fearfulness are inversely related to rate of avoidance conditioning, but no relationship was observed between escape conditioning and any measures of emotional behavior.

Intelligence and Achievement: the "Jangle Fallacy" Again

KELLEY defines the jangle fallacy (4, p. 64) as ... &dquo;the use of two separate words or expressions covering in fact the same basic situation, but sounding different, as though they were in truth different.&dquo; Despite his demonstration (4, pp. 193-209) that the traits measured by group intelligence tests and school achievement tests overlap about 9o per cent, many psychologists as well as school people go on interpreting intelligence test scores as measures of native capacity primarily, and achievement test scores as measures of native capacity plus school motivation and effectiveness of instruction. The present study contributes one more item of evidence showing that a group intelligence test and selected subtests from a school