Reliability of test scores in nonparametric item response theory

Three methods for estimating reliability are studied within the context of nonparametric item response theory. Two were proposed originally by Mokken (1971) and a third is developed in this paper. Using a Monte Carlo strategy, these three estimation methods are compared with four classical lower bounds to reliability. Finally, recommendations are given concerning the use of these estimation methods.The authors are grateful for constructive comments from the reviewers and from Charles Lewis.     

On structural equation modeling with data that are not missing completely at random

A general latent variable model is given which includes the specification of a missing data mechanism. This framework allows for an elucidating discussion of existing general multivariate theory bearing on maximum likelihood estimation with missing data. Here, missing completely at random is not a prerequisite for unbiased estimation in large samples, as when using the traditional listwise or pairwise present data approaches. The theory is connected with old and new results in the area of selection and factorial invariance. It is pointed out that in many applications, maximum likelihood estimation with missing data may be carried out by existing structural equation modeling software, such as LISREL and LISCOMP. Several sets of artifical data are generated within the general model framework. The proposed estimator is compared to the two traditional ones and found superior.The research of the first author was supported by grant No. SES-8312583 from the National Science Foundation and by a Spencer Foundation grant. We wish to thank Chuen-Rong Chan for drawing the path diagram.     

Flicker contrast sensitivity in normal and specifically disabled readers

Temporal contrast sensitivity for counterphase flicker was determined for specifically disabled and normal readers to investigate whether the two groups differ in the functioning of their transient systems. In experiment 1, temporal contrast sensitivity was measured over a range of temporal frequencies with a spatial frequency of 2 cycles deg-1. Disabled readers were less sensitive than the control subjects at all temporal frequencies. In experiment 2, temporal contrast sensitivity was measured at a temporal frequency of 20 Hz over a range of spatial frequencies. Disabled readers were less sensitive than the controls at all spatial frequencies, with the differences between the groups increasing as spatial frequency increased. Both these findings are interpreted as supporting the hypothesis of a transient-system deficit in the visual systems of disabled readers.     

Configurational information in face perception

A new facial composites technique is demonstrated, in which photographs of the top and bottom halves of different familiar faces fuse to form unfamiliar faces when aligned with each other. The perception of a novel configuration in such composite stimuli is sufficiently convincing to interfere with identification of the constituent parts (experiment 1), but this effect disappears when stimuli are inverted (experiment 2). Difficulty in identifying the parts of upright composites is found even for stimuli made from parts of unfamiliar faces that have only ever been encountered as face fragments (experiment 3). An equivalent effect is found for composites made from internal and external facial features of well-known people (experiment 4). These findings demonstrate the importance of configurational information in face perception, and that configurations are only properly perceived in upright faces.     

Relative effectiveness of three stimulus variables for locating a moving sound source

A study is reported in which it is shown that observers can use at least three types of acoustic variables that indicate reliably when a moving sound source is passing: interaural temporal differences, the Doppler effect, and amplitude change. Each of these variables was presented in isolation and each was successful in indicating when a (stimulated) moving sound source passed an observer. These three variables were put into competition (with each indicating that closest passage occurred at a different time) in an effort to determine their relative importance. It was found that amplitude change dominated interaural temporal differences which, in turn, dominated the Doppler effect stimulus variable. The results are discussed in terms of two interpretations. First, it is possible that subjects based their judgements on the potential discriminability of each stimulus variable. However, because the stimuli used involved easily discriminable changes, subjects may instead have based their judgements on the independence of a stimulus variable from different environmental situation conditions. The dominance ordering obtained supports the second interpretation.     

Individual differences and measures of aggression in laboratory studies

After inducing hostility toward a confederate by threatening subjects' self-esteem, subjects were then given the opportunity either to deliver shocks (Experiment 1) or to withhold rewards (Experiment 2) from their confederate. Physiological measures were taken prior to the hostility induction, shortly after the induction and, finally, after the opportunity to aggress. Measures of individual differences relating to aggressive behavior were also considered. While the experimental manipulation was the best predictor in the ‘withholding rewards’ condition, measures of dissipation-rumination tendencies and emotional vulnerability were the best predictors in the ‘shock administration’ condition. In both conditions, systolic blood pressure seemed to reflect differences in arousal as a function of the hostility induction procedure, while subjects in the withhold rewards procedure also showed a decrease in systolic pressure after having an opportunity to aggress toward the confederate. It was concluded that not only is the withholding of rewards a more ethically acceptable procedure than shock administration, but it is also more likely to reflect experimental rather than individual difference affects.     

Dissociation of semantic priming, word and non-word repetition effects by event-related potentials

Two experiments investigated the modulation of event-related potentials (ERPs) by semantic priming and item repetition. In Experiment 1, subjects silently counted occasional non-words against a background of words, a proportion of which were either semantic associates or repetitions of a preceding word. Compared to control items, ERPs to repeated words were distinguished by an early (ca. 200 msec) transient negative-going deflection and a later, topographically widespread and temporally sustained positive-going shift. In contrast, semantically primed words showed a relatively small, topographically and temporally limited positive-going modulation peaking around 500 msec. These data were interpreted as evidence against models of priming and repetition which postulate similar loci for these effects. In Experiment 2, subjects counted occasional words against a background of non-words, some of which were repeated. ERPs to repetitions showed a similar early ERP modulation to that in Experiment 1, and also displayed a later slow positive shift. This latter effect was smaller in magnitude and had a delayed onset in comparison to Experiment 1. It was concluded that the effects of repetition differ as a consequence of whether, prior to their first presentation, items possess a representation in lexical memory.     

Remembering left-right orientation of pictures

In a study of recognition memory for pictures, we observed an asymmetry in classifying test items as "same" versus "different" in left-right orientation: Identical copies of previously viewed items were classified more accurately than left-right reversals of those items. Response bias could not explain this asymmetry, and, moreover, correct "same" and "different" classifications were independently manipulable: Whereas repetition of input pictures (one vs. two presentations) affected primarily correct "same" classifications, retention interval (3 hr vs. 1 week) affected primarily correct "different" classifications. In addition, repetition but not retention interval affected judgments that previously seen pictures (both identical and reversed) were "old". These and additional findings supported a dual-process hypothesis that links "same" classifications to high familiarity, and "different" classifications to conscious sampling of images of previously viewed pictures.     

Multiple regression analysis of twin and sibling data

DeFries and Fulker (1985) employed a multiple regression analysis of twin data in which a cotwin's score is predicted from the proband's score, the coefficient of relationship (R = 1.0 for monozygotic twins and 0.5 for dizygotic twins), and their interaction to yield direct estimates of heritability (h²) and the proportion of variance due to common or shared environmental influences (c²). The purpose of the report is to extend this model to allow for the analysis of data from more than two familial relationships simultaneously. Data from identical twins, fraternal twins, and siblings pairs, in which one member of each pair was selected based on low reading performance, were analyzed. To analyze the data simultaneously, additional coefficients were added to the regression equation in order to differentiate shared environmental influences of twins and siblings. When the model was applied to the combined cognitive data, significant estimates for heritability were obtained in three of the six tests analyzed. The results also indicated a lack of justification for the separate twin and sibling environmental parameters, and a discussion of more parsimonious models is included.