A feasible method for standard errors of estimate in maximum likelihood factor analysis

A jackknife-like procedure is developed for producing standard errors of estimate in maximum likelihood factor analysis. Unlike earlier methods based on information theory, the procedure developed is computationally feasible on larger problems. Unlike earlier methods based on the jackknife, the present procedure is not plagued by the factor alignment problem, the Heywood case problem, or the necessity to jackknife by groups. Standard errors may be produced for rotated and unrotated loading estimates using either orthogonal or oblique rotation as well as for estimates of unique factor variances and common factor correlations. The total cost for larger problems is a small multiple of the square of the number of variables times the number of observations used in the analysis. Examples are given to demonstrate the feasibility of the method.The research done by R. I. Jennrich was supported in part by NSF Grant MCS 77-02121. The research done by D. B. Clarkson was supported in part by NSERC Grant A3109.     

Learning a new human-computer alphabet: the role of similarity and practice

Two purposes motivated this study: (a) to quantify the difficulty in learning various symbols of the alphabet used to enter data into a personal digital assistant (PDA), and (b) to investigate the interaction of item difficulty with practice conditions that promote varying levels of cognitive effort. Levels of compatibility between members of the PDA alphabet and English were quantified through introspective ratings in Experiment 1 and objective performance measures in Experiment 2. Three levels of item compatibility were learned under conditions of proactive or retroactive augmented information in Experiment 3. Contrary to expectations, the item similarity effect did not interact with practice schedules—a retroactive augmented information condition resulted in degraded levels of acquisition performance, but superior retention levels, compared to the proactive condition. These findings are discussed in terms of the relative merits of cognitive effort in skill acquisition.     

Multiple,speeded assessments: Initial evidence on subgroup differences and applicant perceptions

Recently, shorter assessments have emerged as potential alternatives for more resourceful traditional selection approaches. Multiple, speeded assessments (MSAs) represent such an alternative. In MSAs, candidates participate in a large number of short (a maximum of 5 min), behavioral simulations in which they face a variety of job situations. Initial psychometric evidence on the validity of MSAs is promising. Yet, validity represents only one piece of evidence. It is not known whether MSAs disadvantage specific subgroups, which may inhibit diversity. There is also no information on candidates' experience of going through an MSA, which is pivotal for the attractiveness of the organization's selection process. Therefore, this study investigates an MSA in terms of subgroup differences (gender and nationality) and applicant perceptions. Master of Business Administration (MBA) students (N = 96) proceeded through 18 short role-plays sampling junior management situations. Score differences between men and women were negligible. Yet, there were large score differences between national citizens and foreigners. There was no evidence for predictive bias for nationality, though. Of the applicant reaction measures, interpersonal treatment perceptions contributed most to overall fairness perceptions. These findings add to the evidence in support of MSAs, while also stressing to remain vigilant for potential score differences among subgroups.     

Software informed consent: Docete emptorem, not caveat emptor

Should software be sold “as is”, totally guaranteed, or something else? This paper suggests that “informed consent”, used extensively in medical ethics, is an appropriate way to envision the buyer/developer relationship when software is sold. We review why the technical difficulties preclude delivering perfect software, but allow statistical predictions about reliability. Then we borrow principles refined by medical ethics and apply them to computer professionals.     

A computer touch-screen apparatus for training visual discriminations in rats.

We describe an operant conditioning apparatus that uses computerized touch-screen technology and is designed for the versatile and highly controlled testing of rats in a potentially wide variety of behavioral paradigms. Although computer-controlled touch-screen systems have been developed for use with pigeons, monkeys, and humans, analogous technologies and methods have not yet been developed for rats. The development of a touch-screen system for rats could enhance the efficiency of behavioral research with rats, and may offer a unique tool for studying animal learning. In the first test of the utility of the apparatus, 3 Sprague-Dawley rats learned to activate the touch screen only after the touch-screen panel was made slightly movable. These animals then learned to discriminate visual stimuli presented on the computer monitor, but only after the food magazine and pellet dispenser were moved to the rear of the chamber opposite the stimulus display and response window. In a test of the utility of the modified apparatus, 6 Long-Evans rats learned to activate the touch screen and learned one of three different simple discriminations using computer-generated, visually presented stimuli. A basic method for training rats to activate the computer touch screen and for visual discrimination training is described. Results show that rats learned to activate the touch screen and discriminate visual stimuli presented on a computer monitor. Potential applications and advantages of the touch-screen-equipped rat operant conditioning chamber are discussed.     

Unique ethical problems in information technology

A distinction is made between moral indoctrination and instruction in ethics. It is argued that the legitimate and important field of computer ethics should not be permitted to become mere moral indoctrination. Computer ethics is an academic field in its own right with unique ethical issues that would not have existed if computer technology had not been invented. Several example issues are presented to illustrate this point. The failure to find satisfactory non-computer analogies testifies to the uniqueness of computer ethics. Lack of an effective analogy forces us to discover new moral values, formulate new moral principles, develop new policies, and find new ways to think about the issues presented to us. For all of these reasons, the kind of issues presented deserve to be addressed separately from others that might at first appear similar. At the very least, they have been so transformed by computing technology that their altered form demands special attention.     

The accuracy of automated and clinical detection of brain damage and lateralization in neuropsychology

The validity of both computer programs and clinical judgment in neuropsychology for determining the existence and lateralization of brain damage is reviewed. Computerized interpretation in neuropsychology, after a propitious beginning, was largely abandoned due to severe criticism, essentially based on only three studies. Only one of these studies compared clinical judgment with computer programs. A thorough examination of the literature located many more studies assessing the accuracy of computer programs, clinical judgment, and discriminant analysis. When reviewed, these studies found that the computer programs, especially the Neuropsychological Key, were quite accurate though not as accurate as clinical judgment. Computer programs and especially the Lateralization Index are potentially as accurate as expert clinical judgment. The rationale related to computer programs is also discussed. This includes the implications of impairment, criterion adequacy, and methods used in designing the neuropsychological Key and the Lateralization Index. Since computer programs are completely reliable across studies, they can be used to examine the differences between sample populations and criterion accuracy. Factors contributing to reduced accuracy in both clinical judgment and computer programs are also explicated.     

FRANK TIPLER'S PHYSICAL ESCHATOLOGY

Abstract. Frank L. Tipler's book The Physics of Immortality is a striking attempt by a scientist to resolve the conflict between theology and science on the basis of a physicalist position that identifies theology as a branch of physics, and that calculates God “in exactly the same way as physicists calculate the characteristics of electrons.” Tipler's work may be seen as a scientistic myth, and its critique is organized around the three basic characteristics of such myths: (1) it is illogical in that it argues as if physics were in fact metaphysics; (2) it is grim in that its glorification of technology is insensitive to ethical issues; (3) it is meaningless in that its espousal of a strong theory of artificial intelligence empties concrete personal histories by subsuming them under abstractions that distort our understandings both of God and of resurrection.     

THE OMEGA POINT AS ESCHATON: ANSWERS TO PANNENBERG'S QUESTIONS FOR SCIENTISTS

Abstract. I present an outline of the Omega Point theory, which is a model for an omnipresent, omniscient, omnipotent, evolving, personal God who is both transcendent to spacetime and immanent in it, and who exists necessarily. The model is a falsifiable physical theory, deriving its key concepts not from any religious tradition but from modern physical cosmology and computer science; from scientific materialism rather than revelation. Four testable predictions of the model are given. The theory assumes that thinking is a purely physical process of the brain, and that personality dies with the brain. Nevertheless, I show that the Omega Point theory suggests a future universal resurrection of the dead very similar to the one predicted in the Judeo-Christian-Islamic tradition. The notions of "grace" and the "beatific vision" appear naturally in the model.