Robust estimation of ability in the Rasch model

Estimating ability parameters in latent trait models in general, and in the Rasch model in particular is almost always hampered by noise in the data. This noise can be caused by guessing, inattention to easy questions, and other factors which are unrelated to ability. In this study several alternative formulations which attempt to deal with these problems without a reparameterization are tested through a Monte Carlo simulation. It was found that although no one of the tested schemes is uniformly superior to all others, a modified jackknife stood out as the best one in general, it was also super efficient (more efficient than the asymptotically optimal estimator) for tests with forty or fewer items. It is proposed that this sort of jackknifing scheme for estimating ability be considered for practical work.This research was funded through a grant from the Law Enforcement Assistance Administration (78-NI-AX-0047) to the Bureau of Social Science Research, Howard Wainer, Principal Investigator. We would like to thank Ronald Mead, Anne Morgan and James Ramsay for kind, generous, and invaluable help at various stages of the project.     

Classes,collections, and distinctive features: Alternative strategies for solving inclusion problems

This study analyzed the strategies that children ages 5 through 8 years used on two modified versions of Inhelder and Piaget's (The early growth of logic in the child. New York: Norton, 1964) class inclusion task. In two experiments, children were tested on Wilkinson's (Cognitive Psychology, 1976, 8, 64–85) “percept” inclusion task in which distinctive features marked both supraordinate and subclasses. It was hypothesized that children who fail standard Piagetian inclusion tasks succeed on the “percept” task by counting and comparing mutually exclusive features rather than using features as markers for classes and subclasses. The hypothesis was supported by children's performances on “percept” tasks in which solutions based on feature counting conflicted with solutions based on consideration of class inclusion relations. In two other experiments, children answered part-whole and part-part comparison questions in which both terms were described as classes and/or subclasses, or in which one of the two terms was described as a collection (e.g., a bunch of grapes). These experiments contrasted Markman and Seibert's (Cognitive Psychology, 1976, 8, 561–577) “organization” hypothesis that the greater psychological integrity of collections facilitates reasoning on part-whole comparison problems with the hypothesis that the faciltative effect results from the “large number” connotation of collective nouns. Results on collection problems in which parts were described as collections supported the “large number” hypothesis. Results were discussed in terms of their implications for Piaget's theory.     

Unequal Power Relationships Within a Cooperative or Corn petit ive Context

Sixty undergraduates negotiated an exchange of resources in dyads. They had either high or low power relative to the other and were within a competitive or cooperative context. Results support the hypothesis that these contexts affect unequal power persons' interaction and relationship. Compared to those in the cooperative context, high and lower power participants in the competitive context suspected each other, refused to exchange resources, developed negative attitudes toward each other, and associated power with control. In addition, low power participants in the competitive context indicated considerable insecurity about the interaction. These results highlight that previous literature has tended, often implicitly, to consider power and unequal power relationships within a competitive context. The dynamics proposed by this literature that inequality of power undermines negotiations and that power refers to the control of another were found to be much stronger within a competitive context than a cooperative one. Whereas the emphasis of the literature has been on power equalization, findings suggest that placing power differences in a cooperative con text can contribute to effective working relationships.     

The inferiority game: Perceptions and behavior

Other-sex perceptions and actual reported behavior of 318 college males and 287 college females were compared for four dating situations. The average degree of inaccuracy (disregarding direction) for males was 29.0%, while for women it was 12.5%. Conclusions support other research indicating superior role-taking ability on the part of females. The issues of cultural transmission and pluralistic ignorance are discussed as they affect male and female interaction. Important research implications for human relations and the family in particular are noted.     

Determinants of satisfaction with specific job facets: A test of Locke's model

Survey results supported Locke's (1969, 1976) contention that facet job satisfaction is a function of three basic determinants: 1) facet amount; 2) wanted amount; and 3) facet importance. We found several three-way interactions that were generally consistent with Locke's model. Specifically, facet satisfaction was highest when employees wanted a large amount of the facet in question, were currently receiving a large amount, and felt that the facet was personally important. Conversely, satisfaction was lowest when employees wanted a large amount of the facet, were currently receiving only a small amount, and felt that the facet was personally important. When job facets were viewed as unimportant to employees, however, the amount of facet desired and currently possessed had little impact on satisfaction. Implications, limitations, and directions for research are discussed.Bob Rice passed away recently. He will be sorely missed.     

Do learning-disabled children exhibit peripheral deficits in selective attention? An analysis of eye movements during reading

It has been suggested that deficits in selective attention play a functional role in the learning and behavior problems of children diagnosed as learning- disabled. In the present study, peripheral and central aspects of selective attention were distinguished and peripheral aspects were examined. The attending eye movements during reading of 12 fifth-grade learning-disabled children who read at the third-grade level were compared with 12 fifth-grade children who read at the fifth-grade level (grade placement controls) and 12 third-grade children who read at the third-grade level (reading level controls). The learning-disabled children did not differ from their reading level controls on any of W measures but showed a pattern of eye movements that was generally slower and less smooth than their grade placement controls. While these patterns might be considered to be immature relative to normal readers of the same age, they are not indicative of peripheral deficits in selective attention. Males, however, did exhibit more variability in duration of fixations and made more regressive eye movements than did females.     

Cerebral laterality and verbal-performance discrepancies in intelligence

The cerebral laterality of children with various configurations of verbal-performance discrepancies was inferred with an objective measure of lateral preference. Specifically, Verbal and Performance IQ scores of the WISC-R were used to divide 90 learning-disabled children into three equal groups: (a) Verbal significantly higher than Performance, (b) Performance significantly higher than Verbal, and (c) Verbal not significantly different from Performance. Analysis of the scores on the Laterality Preference Schedule showed children in the Verbal > Performance group were significantly more bilateral in their preference patterns than either of the other groups. Data on familial laterality showed this Verbal > Performance group to also have parents who were significantly more mixed in their preference patterns than either the VSIQ < PSIQ or VSIQ = PSIQ groups. The results were interpreted as lending support to the notion of competition antagonism between cortical hemispheres and a possible genetic relationship.     

One-trial overshadowing

Two experiments on conditioned suppression in rats examined the extent to which conditioning to a tone was overshadowed by the presence of a light. When animals received four conditioning trials, a light of moderate intensity significantly overshadowed the tone. On the first trial of conditioning, however, the light had no effect on conditioning to the tone, unless (Experiment I) its intensity was increased, or (Experiment II) it occurred in closer temporal contiguity to reinforcement than the tone. The results establish that overshadowing can occur on the first trial of conditioning, contrary to the predictions of several theories, but suggest that this may be for different reasons from those operating to produce overshadowing over a series of conditioning trials.     

Vicious-circle behavior in the Y maze

Laboratory rats were given escape training in both arms of a Y maze followed by extinction in one of four conditions: regular (no shock), punished (both arms containing a shock mid-segment), choice, or forced (one arm containing a shock mid-segment, one arm shock free). Subjects in the punished and forced conditions required more extinction trials than those in the regular extinction condition. The choice condition was intermediate and not significantly different from the others. Choice subjects avoided the shock by choosing the safe alley early and consistently during extinction, but some continued to run for many trials without further punishment. It was suggested that the number of punished trials early in extinction is related to persistence of running, and the results were discussed in relation to the Mowrer-Brown and the discrimination hypotheses.