Provides a suitable test for the hypothesis rho = O and establishes nomograms by which general users of R can set confidence intervals on rho. (Author/RC)

Marascuilo (1966) proposes a chi square analog of Scheffe's multiple comparisons procedure. With respect to pairwise comparisons among k independent correlation coefficients, a multiple range procedure will produce lower critical values than the corresponding chi square values, thereby increasing the power of the resulting tests. Basis of the…

Equations are derived for product correlation between variables derived from two or fewer original variables. Index correlation and spurious index correlation are briefly discussed. Exact equations for product correlation are derived, and spurious product correlation is demonstrated under various conditions. (RC)

Results suggest that different groups of judges do set similar examination standards when using the same procedure, and that the average of individual judgments does not differ significantly from group consensus judgments. Significant differences were found, however, between the standards set by the two procedures employed. (RC)

A method of analysis is developed which accounts for whatever variance is present in the responses to test items, additive, configural or the two combined. The method is applied to successive column matrices which report frequencies or endorsements by categories of subjects to answer alternative sets of test items. Approaches are included which…

Using the concepts of conditional probability, conditional expectation, and conditional independence, the main results of the classical test theory model can be derived in a very few steps with minimal assumptions. The present effort explores the possibility that present classical test theories can be further condensed. (Author/RC)

Investigates the Tukey statistic for the empirical probability of a Type II error under numerous parametric specifications defined by Cohen (1969) as being representative of behavioral research data. For unequal numbers of observations per treatment group and for unequal population variancies, the Tukey test was simulated when sampling from a…

A test is developed for hypotheses about the grand mean in the analysis of variance, using the known relationship between the t distribution and the F distribution with 1 df (degree of freedom) for the numerator. (Author/RC)

Investigates the question of under what conditions can one infer that some observed relationship between two empirical variables reflects more than that they are both indicants of the same construct. (RC)

Illustrates a multivariate approach to item analysis. Previous formulation is extended by investigating techniques simultaneously taking into account scale variance with the goal of reducing the average correlation between scales. Study examines problems in determining optimum values for combinations of item parameters selected for personality…

Illustrates the use of multiple regression analysis for computing conditional probabilities of occurrence of events based on the functional relationship between a dependent response variate and one or more independent predictors. (RC)

Some of the ramifications of the linear order assumption underlying much of measurement theory are discussed. Ordering theory is then described as an alternative deterministic measurement model that does not make the linear order assumption. (Author/RC)

Investigates the relation between factor score estimates, principal component scores, and image scores. The three methods compared are maximum likelihood factor analysis, principal component analysis, and a variant of rescaled image analysis. (RC)

Adults read a prose passage and responded to passages based on it which were either true or false and were phrased either affirmatively or negatively. True negatives yielded most errors, followed in order by false negatives, true affirmatives, and false affirmatives. (Author/RC)

Describes a computer program for calculating the power of the F-test. Approach is based upon two independent approximations-- a normalization of the non-central F distribution and an integration of the normal distribution. Comparison of the calculated values of power with exact values revealed a high degree of accuracy. (Author/RC)