There are at least two situations in which the behavioral scientist wishes to transform uniformly distributed data into normally distributed data: (1) In studies of sampling distributions where uniformly distributed pseudo-random numbers are generated by a computer but normally distributed numbers are desired; and (2) In measurement applications…

Computational formulae for the following three tests of significance, useful in the interrupted time series design, are given: (1) a "t" test (Mood, 1950) for the significance of the first post-change observation from a value predicted by a linear fit of the pre-change observations; (2) an "F" test (Walker and Lev, 1953) of the…

In a pilot study of college students' understanding of mathematical induction, the author asked 2,320 students and 152 faculty members to determine whether eight statements were true or false. The difficulty of the statements was analyzed for subgroups of the population. (SD)

Presents two models for statistically analyzing occupational mobility. The models show how the destination distribution is shifted for each shift of one or more steps on the scale of occupational origins. (Author/AV)

The controversy surrounding regression methods for unbalanced factorial designs is addressed. The statistical hypotheses being tested under the various methods, as well as salient issues in the use of these methods, are discussed. The use of statistical computer packages is also discussed. (Author/JKS)

Estimating the weight of large pumpkins before harvest presents an opportunity for applying several diverse mathematical topics. A model that allows an estimation by easy tape measurement is derived. (MP)

A comparative study of the cost effectiveness of retrospective manual and on-line bibliographic searching is described. Individual task times for question analysis, searching, photocopying, shelving and output distribution were measured, and component costs for labor, information, reproduction, physical space and telecommunications calculated. The…

Path analyses of two multitrait-multimethod matrices are used as examples of the kind of information afforded by application of the technique. It is concluded that the technique should be of considerable aid to researchers who want to evaluate the convergent and discriminant validity of their measures. (Author/CTM)

A goodness of fit test presented by Andersen (EJ 143 939) is shown to be incorrect. The correct test is described and a re-analysis of Andersen's data is provided. (Author/CTM)

Graham Douglas' claims (TM 503 496) that the X2-test statistics of the paper, "Paired Comparisons with Individual Differences" (EJ 143 939), are incorrect are acknowledged to be justified (Author/CTM)

This model is designed to determine the optimal resource mix necessary to meet a set of educational goals within a set of technical and physical resource constraints. (Author/IRT)

The mathematical description of recidivism is shown to be improved by using a limited or asymptotic growth curve rather than a linear relationship over time. The proposed curve for failure rate may also be applied to other programs for behavioral change. (CTM)

Two procedures for Rasch, sample-free item calibration are reviewed and compared for accuracy. The theoretically ideal "conditional" procedure is impractical for more than fifteen items. The more practical but biased "unconditional" procedure is discussed in detail. (Author/JKS)

The problems of two way analysis of variance designs with unequal and disproportionate cell sizes are discussed. A variety of solutions are discussed and a new solution is presented. (JKS)

The action in a British sporting event (bumping races) was used to motivate a simple method of computing the correlation between starting and finishing positions. The method is generalized to other situations. (SD)