A comparison of factor analysis programs in SPSS,BMDP, and SAS

Factor analysis programs in SAS, BMDP, and SPSS are discussed and compared in terms of documentation, methods and options available, internal logic, computational accuracy, and results provided. Some problems with respect to logic and output are described. Based on these comparisons, recommendations are offered which include a clear overall preference for SAS, and advice against general use of SPSS for factor analysis.     

A comparison of the efficiency and accuracy of BILOG and LOGIST

Comparisons are made between BILOG version 2.2 and LOGIST 5.0 Version 2.5 in estimating the item parameters, traits, item characteristic functions (ICFs), and test characteristic functions (TCFs) for the three-parameter logistic model. Data analyzed are simulated item responses for 1000 simulees and one 10-item test, four 20-item tests, and four 40-item tests. LOGIST usually was faster than BILOG in producing maximum likelihood estimates. BILOG almost always produced more accurate estimates of individual item parameters. In estimating ICFs and TCFs BILOG was more accurate for the 10-item test, and the two programs were about equally accurate for the 20- and 40-item tests.I am grateful to Robert J. Mislevy, Martha L. Stocking, and Marilyn S. Wingersky for many helpful comments on an earlier version of this paper. I would also like to thank Hamid Kamrani and Bongmyoung Park for getting LOGIST and BILOG running and keeping them running under changing computer systems at CTB/McGraw-Hill.     

Psychological distance to reward: A human replication

Choice behavior in college students was examined in two experiments using the concurrent-chains procedure. In both experiments, the concurrent chains were presented on a microcomputer in the form of an air-defense game in which subjects used two radar systems to detect and subsequently destroy enemy aircraft. Access to one of two radar systems was controlled by a pair of independent concurrent variable-interval 60-s schedules with a 4-s changeover delay always in effect. In the terminal link, the appearance of an enemy aircraft was determined by a pair of differentially segmented fixed-time schedules (Experiment 1) or fixed-interval schedules (Experiment 2) of equal overall duration. In Experiment 1, the terminal-link duration was either 20 s or 40 s, and subjects preferred the unsegmented to the segmented intervals. In Experiment 2, the duration was either 10 s or 60 s, and the reinforcement contingencies required responding during the terminal link. Prior to the reinstatement of the initial link, subjects estimated the duration of the terminal-link schedule. Segmentation affected choice in the 60-s conditions but not in the 10-s ones. Preference for the unsegmented schedule was correlated with an overestimation of the durations for the segmented schedules. These results replicated those found in animal experiments and support the notion of increasing the psychological distance to reward by segmenting a time-based schedule of reinforcement.     

Statistics as psychometrics

In this paper, modern statistics is considered as a branch of psychometrics and the question of how the central problems of statistics can be resolved using psychometric methods is investigated. Theories and methods developed in the fields of test theory, scaling, and factor analysis are related to the principle problems of modern statistical theory and method. Topics surveyed include assessment of probabilities, assessment of utilities, assessment of exchangeability, preposterior analysis, adversary analysis, multiple comparisons, the selection of predictor variables, and full-rank ANOVA. Reference is made to some literature from the field of cognitive psychology to indicate some of the difficulties encountered in probability and utility assessment. Some methods for resolving these difficulties using the Computer-Assisted Data Analysis (CADA) Monitor are described, as is some recent experimental work on utility assessment.1980 Psychometric Society presidential address.I am indebted to Paul Slovic and David Libby for valuable consultation on the issues discussed in this paper and to Nancy Turner and Laura Novick for assistance in preparation.Research reported herein was supported under contract number N00014-77-C-0428 from the Office of Naval Research to The University of Iowa, Melvin R. Novick, principal investigator. Opinions expressed herein reflect those of the author and not those of sponsoring agencies.     

Effects of computer‐based training on procedural modifications to standard functional analyses

Few studies have evaluated methods for training decision‐making when functional analysis data are undifferentiated. The current study evaluated computer‐based training to teach 20 graduate students to arrange functional analysis conditions, analyze functional analysis data, and implement procedural modifications. Participants were exposed to training materials using interactive software during a 1‐day session. Following the training, mean scores on the posttest, novel cases probe, and maintenance probe increased for all participants. These results replicate previous findings during a 1‐day session and include a measure of participant acceptability of the training. Recommendations for future research on computer‐based training and functional analysis are discussed.     

The development of computer ethics: Contributions from business ethics and medical ethics

In this essay, we demonstrate that the field of computer ethics shares many core similarities with two other areas of applied ethics, Academicians writing and teaching in the area of computer ethics, along with practitioners, must address ethical issues that are qualitatively similar in nature to those raised in medicine and business. In addition, as academic disciplines, these three fields also share some similar concerns. For example, all face the difficult challenge of maintaining a credible dialogue with diverse constituents such as academicians of various disciplines, professionals, policymakers, and the general public, Given these similarities, the fields of bioethics and business ethics can serve as useful models for the development of computer ethics. A version of this paper was presented at ETHICOMP98, the Fourth International Conference on Ethical Issues of Information Technology, March 25–27, 1998, Erasmus University, the Netherlands. Kenman Wong, Ph.D., is an Associate Professor of Business Ethics; Gerhard Steinke, Ph.D., is Professor of Management and Information Systems. Both authors are at Seattle Pacific University's School of Business and Economics.     

Classification of vowels and consonants by individuals with moderate mental retardation: development of arbitrary relations via match-to-sample training with compound stimuli.

This study explored whether an identity-matching-based stimulus equivalence procedure could be used to teach vowel and consonant stimulus classes to 2 adolescent females with moderate mental retardation. Delayed match-to-sample trials presented a compound sample stimulus consisting of printed letters and a spoken word (“vowel” or “consonant”). The correct comparison stimulus matched only one of the letters in the compound sample. Subsequently, test trials assessed whether arbitrary relations had formed among the individual stimuli from each compound sample and whether stimuli from different compound samples had merged into larger stimulus classes. Both participants acquired five-member classes of vowel and consonant stimuli, which subsequently generalized to vocal classification and to identification in the context of four-letter words. Follow-up tests showed that the generalized performances remained intact after 6 weeks. These procedures suggest an economical approach to stimulus class development.     

Generalization of cross-modal stimulus equivalence classes: operant processes as components in human category formation.

This study employed a stimulus-class rating procedure to explore whether stimulus equivalence and stimulus generalization can combine to promote the formation of open-ended categories incorporating cross-modal stimuli. A pretest of simple auditory discrimination indicated that subjects (college students) could discriminate among a range of tones used in the main study. Before beginning the main study, 10 subjects learned to use a rating procedure for categorizing sets of stimuli as class consistent or class inconsistent. After completing conditional discrimination training with new stimuli (shapes and tones), the subjects demonstrated the formation of cross-modal equivalence classes. Subsequently, the class-inclusion rating procedure was reinstituted, this time with cross-modal sets of stimuli drawn from the equivalence classes. On some occasions, the tones of the equivalence classes were replaced by novel tones. The probability that these novel sets would be rated as class consistent was generally a function of the auditory distance between the novel tone and the tone that was explicitly included in the equivalence class. These data extend prior work on generalization of equivalence classes, and support the role of operant processes in human category formation.     

Measuring nominal scale agreement between a judge and a known standard

Two designs for comparing a judge's ratings with a known standard are presented and compared. Design A pertains to the situation where the judge is asked to categorize each ofN subjects into one ofr (known) classes with no knowledge of the actual number in each class. Design B is employed when the judge is given the actual number in each class and is asked to categorize the individuals subject to these constraints. The probability distribution of the total number of correct choices is developed in each case. A power comparison of the two procedures is undertaken.