Generating multiple samples of multivariate data with arbitrary population parameters

A method of generating any number of score and correlation matrices with arbitrary population parameters is described. EitherZ scores or stanines are sampled from a normal population to represent factor scores by an IBM 1620 program. These are converted to variates from a population with an a priori factor structure. The effectiveness of the method is illustrated from research data. Some further modifications and uses of the method are discussed.     

Generating possibilities

Our knowledge of the most basic alternative possibilities can be thought of as generated recursively from what we know about the actual world. But what are the generating principles? According to one view, they are recombinational: roughly, alternative possibilities are generated by “patching together” parts of distinct worlds or “blotting out” parts of worlds to yield new worlds. I argue that this view is inadequate. It is difficult to state in a way that is true and non-trivial, and anyway fails to account for our knowledge that there might have been other things, properties, relations, and combinations of these than there actually are. I sketch and defend an alternative view based on the distinction between determinable and determinate properties: roughly, alternative possibilities are generated by “intra-determinable” variation, variation from one determinate to another of the same determinable.     

Linear structural equations with latent variables

An interdependent multivariate linear relations model based on manifest, measured variables as well as unmeasured and unmeasurable latent variables is developed. The latent variables include primary or residual common factors of any order as well as unique factors. The model has a simpler parametric structure than previous models, but it is designed to accommodate a wider range of applications via its structural equations, mean structure, covariance structure, and constraints on parameters. The parameters of the model may be estimated by gradient and quasi-Newton methods, or a Gauss-Newton algorithm that obtains least-squares, generalized least-squares, or maximum likelihood estimates. Large sample standard errors and goodness of fit tests are provided. The approach is illustrated by a test theory model and a longitudinal study of intelligence.This investigation was supported in part by a Research Scientist Development Award (KO2-DA00017) and a research grant (DA01070) from the U. S. Public Health Service.     

Generating complex waveforms

A wide variety of complex waveforms can be generated by approximating the desired analog waveform from an array of digital values. Some basic properties of these digital approximations are discussed in terms of pulse amplitude modulation and sampling theory. The waveforms are generated by transferring the digital values to a digital-to-analog converter followed by a low-pass filter. This usually requires the dedicated use of a computer. We have built a device, incorporating solid state memory, that can store, time, and transfer previously computed digital values, so that a computer is no longer necessary to generate the waveforms. Specifications of the digital-to-analog converter and appropriate settings of the filter are discussed, along with a simplified procedure for calculating waveforms that have line spectra. An adaptation of this procedure enables the device to be used as a high-speed programmable pure-tone source.     

Generating probable consequences

A study was conducted to examine some of the factors that influence peoples' generation of likely consequences associated with a major hypothetical life event, involving the loss of the function of one of one's own limbs. The design included four limbs involving two inclusion relations, thumb-hand and foot-leg, and two viewpoints involving consequences for one-self or for others. The results indicated that subjects generate more numerous consequences for primary than for secondary consequences, more for short-term consequences than for medium- or long-term ones, more for an inclusive limb than for an included limb (e.g., hand vs thumb), and more pertaining to one-self than to others. Individual differences in the generation of consequences were not associated with prior experience involving functional loss of the limbs tested or with general differences in academic performance. The results permit inference about some of the mental processes that accompany consequence generation.     

Generating nonsymbolic number stimuli

Studies investigating nonsymbolic numbers (e.g., dot arrays) are confronted with the problem that changes in numerosity are always accompanied by changes in the visual properties of the stimulus. It is therefore debated whether the visual properties of the stimulus rather than number can explain the results obtained in studies investigating nonsymbolic number processing. In this report, we present a program (available at ; note that the program is designed to work with the Psychophysics Toolbox in MATLAB) that exports information about the visual properties of stimuli that co-vary with number (area extended, item size, total surface, density, and circumference). Consequently, insight into the relation between the visual properties of the stimulus and numerical distance can be achieved, and post hoc analyses can be conducted to directly reveal whether numerical distance or (some combinations of) the visual properties of a stimulus could be the most likely candidate underlying the results. Here, we report data that demonstrate the program’s usefulness for research on nonsymbolic number stimuli.     

Personality and environmental variables associated with dental anxiety

94 adult dental patients' anxiety was assessed. Scores on Spielberger, et al.'s inventory showed attractiveness of the waiting room, age, and Edwards' Harmavoidance were significantly related to anxiety.     

Structural equation models with continuous and polytomous variables

A two-stage procedure is developed for analyzing structural equation models with continuous and polytomous variables. At the first stage, the maximum likelihood estimates of the thresholds, polychoric covariances and variances, and polyserial covariances are simultaneously obtained with the help of an appropriate transformation that significantly simplifies the computation. An asymptotic covariance matrix of the estiates is also computed. At the second stage, the parameters in the structural covariance model are obtained via the generalized least squares approach. Basic statistical properties of the estimates are derived and some illustrative examples and a small simulation study are reported.This research was supported in part by a research grant DA01070 from the U. S. Public Health Service. We are indebted to several referees and the editor for very valuable comments and suggestions for improvement of this paper. The computing assistance of King-Hong Leung and Man-Lai Tang is also gratefully acknowledged.     

The abstraction of arbitrary categories

Bransford and Franks (1971) showed that people retain the ideas expressed in linguistic input, rather than retaining the individual sentences. The extent to which this phenomenon reflects organizational’ processes based on co-occurrence information was investigated by using Bransford and Franks’ procedure but substituting arbitrarily chosen groups of words for sentences. The subjects gave higher recognition ratings to large new groups (analogous to complex sentences) than to small new groups (analogous to simpler ones), so long as the size of the group was in the range normally presented. They also showed the same ordering for old groups; but they rated old groups higher than new groups within each group size. Hierarchical recognition response structure can be established without an ideational basis, and it does not require complete absence from memory of original input items.     

Natural deduction and arbitrary objects

Journal of Philosophical Logic -     

The association of attitudes toward stuttering with selected variables

The Stuttering Inventory, designed to assess knowledge about and identify attitudes toward stuttering was completed by 203 subjects. The relative contributions of age, gender, years of education, knowledge about stuttering, number of people who stutter personally known, number of stuttering courses completed, and possession of the Certificate of Clinical Competence in speech-language pathology towards the prediction of attitudes toward people who stutter, their parents, and the efficacy of therapy were examined. Possessing the Certificate of Clinical Competence was the most reliable predictor of attitude scores. The limitations of this study and related research efforts using paper and pencil assessments of attitudes toward stuttering are discussed.