The Efficacy of the General Aptitude Test Battery in Allocating Prisoners to Work and Trade Training Courses

Ninety trainees at Deerbolt Borstal, County Durham, were randomly assigned to four groups for placement on trade training courses and work. The first group was assigned on the basis of the General Aptitude Test Battery and a brief interview about their interests; the second group on the basis of the General Aptitude Test Battery; the third group on the interview about their interests; whilst the fourth group was assigned randomly to existing vacancies. A 16-week follow-up showed significant differences across the groups in terms of successfully completing placements, the General Aptitude Test Battery being the crucial factor in allocating individuals successfully. Ratings of trainee performance from supervisors taken after two weeks did not, however, show significant differences across the groups.     

Effects of the menstrual cycle on vibrotactile sensitivity

Psychophysical thresholds for the detection of vibration delivered to the thenar eminence of the right hands of young males and females were measured every other day for 30–34 days. The frequency of the vibratory stimulus was either 15 or 250 Hz. The sinusoidal stimuli applied through a 3.0-cm² contactor were 700 msec in duration and had rise-fall times of 25 msec. Stimulus amplitude was measured with a calibrated accelerometer mounted on the moving element of the vibrator. Data were plotted as a function of successive days in the menstrual cycle. Thresholds for detecting the 15-Hz stimulus did not change significantly, whereas thresholds for detecting the 250-Hz stimulus varied significantly over the menstrual cycle. The 250-Hz threshold became progressively lower as subjects approached the onset of menstruation. After the onset of menstruation, the 250-Hz threshold gradually increased, reaching a maximum value approximately 12 or 13 days later. Soon after the 12th or 13th day of the cycle, the threshold again began to decline and continued to decline until the onset of the next menstrual cycle. The thresholds of women taking birth control pills did not change systematically over the testing period.     

The similarities between cognitive dissonance and guilt: Confession as a relief of dissonance

It is proposed that dissonance is similar to the feeling of guilt. If they are in fact analogous, and confession relieves guilt, then one implication is that confession would relieve dissonance. This experiment tested this prediction using a standard induced compliance paradigm. Forty-five undergraduate subjects were assigned to either a no-choice condition, a choice-confession condition, or a choice-no-confession condition. The results indicate that confession does relieve dissonance. Further, as predicted, a negative correlation was found between attitudinal shift and guilt. The implications of the findings concerning the mechanisms of confession are discussed.     

Causes of young children's confusion of same and different

Three experiments are reported that explore 3-year-olds' and adults' understanding of the words, same and different. In the first, 3-year-olds selected a bead that was “the same color as” or “a different color from” a target bead. In the second, 3-year-olds selected a bead that was “the same in some way as” or “different in some way from” a target bead. Contrary to results reported previously, the majority of consistent responders chose a bead identical with the target in response to the different instruction in both experiments. The rates of both incorrect different responding and incorrect same responding were greater in Experiment II than in Experiment I. In Experiment III, adults chose objects that were “the same as” or “different from” a target; unlike the children, they never chose a target-identical object in response to different instructions. It is argued that children and adults treat same and different differently, and that children's task performance is influenced by three factors: semantic, pragmatic, and nonlinguistic.     

The relationship between treatment exposure times and changes in stuttering frequency during contingent stimulation

Four adult stutterers were exposed to 60 consecutive minutes of verbal contigent stimulation. In each session subjects spoke until a stuttering frequency stability criterion was met (baserate) followed by 60 min of exposure to the word wrong presented contingently upon each instance of stuttering. The data were analyzed according to changes in stuttering which occured following 20, 40, and 60 min of treatment exposure. The results revealed that stuttering frequency was reduced for all subjects during all three 20-min segments of the treatment condition. The only significant reductions in stuttering frequency, however, occured within the first 20 min of treatment.     

Changing expectations: A longitudinal study of community attitudes toward a nuclear power plant

Initial and 5-year follow-up interviews were conducted with 213 residents of the host community for a nuclear power plant. The purpose was to determine possible changes in attitudes toward the plant and expectations about potential outcomes associated with construction. Large negative changes in attitudes toward the plant were noted and were accompanied most notably by decreased expectations of positive outcomes. The structure of the expectations remained essentially stable over the 5-year period. Perceptions of hazards, community disruption, and economic benefits as measured early in construction and during peak construction were found to be the best predictors of acceptance of the nuclear plant. Initial expectations were found to predict overall attitude toward the plant 5 years later. Results were discussed in terms of implications for social impact assessment, large-scale community change, and the predictability of community attitudes toward nuclear power plant construction.     

A Bayesian random effects model for testlets

Standard item response theory (IRT) models fit to dichotomous examination responses ignore the fact that sets of items (testlets) often come from a single common stimuli (e.g. a reading comprehension passage). In this setting, all items given to an examinee are unlikely to be conditionally independent (given examinee proficiency). Models that assume conditional independence will overestimate the precision with which examinee proficiency is measured. Overstatement of precision may lead to inaccurate inferences such as prematurely ending an examination in which the stopping rule is based on the estimated standard error of examinee proficiency (e.g., an adaptive test). To model examinations that may be a mixture of independent items and testlets, we modified one standard IRT model to include an additional random effect for items nested within the same testlet. We use a Bayesian framework to facilitate posterior inference via a Data Augmented Gibbs Sampler (DAGS; Tanner & Wong, 1987). The modified and standard IRT models are both applied to a data set from a disclosed form of the SAT. We also provide simulation results that indicates that the degree of precision bias is a function of the variability of the testlet effects, as well as the testlet design.The authors wish to thank Robert Mislevy, Andrew Gelman and Donald B. Rubin for their helpful suggestions and comments, Ida Lawrence and Miriam Feigenbaum for providing us with the SAT data analyzed in section 5, and to the two anonymous referees for their careful reading and thoughtful suggestions on an earlier draft. We are also grateful to the Educational Testing service for providing the resources to do this research.     

Generation speed in Raven''s progressive matrices test

In this paper, we investigate the role of response fluency on a well-known intelligence test, Raven's Advanced Progressive Matrices (APM) test. Finding rules that govern the items is critical in solving this test. Finding these rules is conceptualized as sampling rules from a (statistical) rule distribution until the correct one is attained. Response fluency is then seen as generation speed, or the speed at which a person generates (samples) rules from this distribution. We develop a test that isolates this speed of sampling variable, and a method to check whether this variable was adequately isolated. The score on this test is then compared with performance on the APM test. It is found that the speed at which people sample from such distributions is an important variable in solving APM items.