It is shown that certain correlation measures suggested by information theory are simple functions of the determinants of the correlation matrices involved, if multivariate normality is assumed. This illuminates and in a sense strengthens Rozeboom's point in [3], that information theoretical statistics have “classical” correlational counterparts.

In the course of developing the minres method of factor analysis the troublesome situation of communalities greater than one arose. This problem—referred to as the generalized Heywood case—is resolved in this paper by means of a process of minimizing the sum of squares of off-diagonal residuals. The resulting solution is superior to the otherwise very efficient original minres method without requiring additional computing time.

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.

A 1-choice structure arises whenever each person in a group chooses exactly one other person according to some criterion. Our purpose is to study the situation in which these choices are made at random. As noted in Harary, Norman and Cartwright [2], such a structure can be represented by a directed graph in which the points represent people and the directed lines their choices. We first describe the shape of such a 1-choice structure, and define its symmetry number. With the help of these properties we are then able to develop and prove a formula which gives a probability of obtaining a given structure in the random choice situation. In order to supply data for these results, we include in the Appendix the diagrams of all 1-choice structures with at most 7 points and the probability of each.

In this paper, maximum-likelihood estimates have been obtained for covariance matrices which have the Guttman quasi-simplex structure under each of the following null hypotheses: (a) The covariance matrix Σ, can be written asTΔT′ + Γ where Δ and Γ are both diagonal matrices with unknown elements andT is a known lower triangular matrix, and (b) the covariance matrix Σ*, is expressible asTΔ*T′ + γI where γ is an unknown scalar. The linear models from which these covariance structures arise are also stated along with the underlying assumptions. Two likelihood-ratio tests have been constructed, one each for the above null hypotheses, against the alternative hypothesis that the population covariance matrix is simply positive definite and has no particular pattern. A numerical example is provided to illustrate the test procedure. Possible applications of the proposed test are also suggested.

This paper presents the results from an investigation of the true probability distributions of the range of rank totals. A procedure for generating an approximation to the true distributions is also given. A comparison of the results of this approximation with an extensive criterion of generated true and sample distributions, and with other approximations is indicated. Accurate estimates of the critical ranges necessary to reach significance at three commonly used alpha levels, where the number of judges and items is less than or equal to sixteen, are presented in tabular form.