Using response times to detect aberrant responses in computerized adaptive testing

A lognormal model for response times is used to check response times for aberrances in examinee behavior on computerized adaptive tests. Both classical procedures and Bayesian posterior predictive checks are presented. For a fixed examinee, responses and response times are independent; checks based on response times offer thus information independent of the results of checks on response patterns. Empirical examples of the use of classical and Bayesian checks for detecting two different types of aberrances in response times are presented. The detection rates for the Bayesian checks outperformed those for the classical checks, but at the cost of higher false-alarm rates. A guideline for the choice between the two types of checks is offered.This study received funding from the Law School Admission Council (LSAC). The opinions and conclusions contained in this paper are those of the authors and do not necessarily reflect the policy and position of LSAC. The authors are most indebted to Wim M. M. Tielen for his computational assistance and to the US Defense Manpower Data Center for the permission to use the ASVAB data set in the empirical examples.     

Effects of alternative reinforcement on human behavior: the source does matter

Competing theories regarding the effects of delivering periodic response-independent reinforcement (more accurately, response-independent points exchanged for money) on a baseline rate of behavior were evaluated in human subjects. Contiguity theory holds that these events decrease target responding because incompatible behavior is adventitiously strengthened when the point deliveries follow target behavior closely in time. Matching theory holds that response-independent points, like any other alternative reinforcer, should reduce target responding. On this view, temporal contiguity between target responding and response-independent point delivery is unimportant. In our experiment, four different responses (moving a joystick in four different directions) were reinforced with points exchangeable for money according to four independent variable-interval schedules. Different schedules of point delivery were then superimposed on these baselines. When all superimposed point deliveries occurred immediately after one of the four responses (the target response), time allocated to target responding increased. When the superimposed point deliveries could be delivered at any time, time allocated to target responding declined and other behavior increased. When superimposed points could never immediately follow target responses, time allocated to target responding decreased further and other behavior or pausing predominated. The findings underscore the contribution of temporal contiguity in the effects of response-independent deliveries of food, money, points, etc.     

A note on the equivalence of the graded response model and the sequential model

This paper concerns items that consist of several item steps to be responded to sequentially. The item scoreX is defined as the number of correct responses until the first failure. Samejima's graded response model states that each steph=1,...,m is characterized by a parameterb _h , and, for a subject with ability, Pr(Xh; )=F(–b _h ). Tutz's general sequential model associates with each step a parameterdh, and it states that Pr(Xh;)= _r ^=1h G(–d _r ). Tutz's (1991, 1997) conjectures that the models are equivalent if and only ifF(x)=G(x) is an extreme value distribution. This paper presents a proof for this conjecture.     

Resistance to change of operant variation and repetition

A multiple chained schedule was used to compare the relative resistance to change of variable and fixed four-peck response sequences in pigeons. In one terminal link, a response sequence produced food only if it occurred infrequently relative to 15 other response sequences (vary). In the other terminal link, a single response sequence produced food (repeat). Identical variable-interval schedules operated in the initial links. During baseline, lower response rates generally occurred in the vary initial link, and similar response and reinforcement rates occurred in each terminal link. Resistance of responding to prefeeding and three rates of response-independent food delivered during the intercomponent intervals then was compared between components. During each disruption condition, initial- and terminal-link response rates generally were more resistant in the vary component than in the repeat component. During the response-independent food conditions, terminal-link response rates were more resistant than initial-link response rates in each component, but this did not occur during prefeeding. Variation (in vary) and repetition (in repeat) both decreased during the response-independent food conditions in the respective components, but with relatively greater disruption in repeat. These results extend earlier findings demonstrating that operant variation is more resistant to disruption than is operant repetition and suggest that theories of response strength, such as behavioral momentum theory, must consider factors other than reinforcement rate. The implications of the results for understanding operant response classes are discussed.     

The ADHD Response-Inhibition Deficit as Measured by the Stop Task: Replication with DSM–IV Combined Type, Extension, and Qualification

Although response inhibition has been proposed as a core element of child attention-deficit hyperactivity disorder (ADHD), the literature is heavily reliant on studies using DSM–III–R diagnostic criteria, older methods of measuring response inhibition, samples of boys, and failing to control thoroughly for comorbid problems—both as diagnoses and as subclinical variation. The present study replicated a deficit in response inhibition in the ADHD combined type (DSM–IV, American Psychiatric Association, 1994) using samples matched on age and sex. The study replicated an effect size of approximately d = .6 in boys with ADHD, and observed an even larger effect size for girls, although the Sex × Group interaction was nonsignificant. Children with ADHD also had problems with response output, shown by variable responding. Excluding comorbid conduct disorder, reading disorder, generalized anxiety disorder, obsessive–compulsive disorder, major depression, and posttraumatic stress disorder from the sample did not alter the results. Correlations indicated that response inhibition was associated with both attentional problems and reading level. Covarying for reading problems did not eliminate the ADHD group effect, but the association of response inhibition with reading clearly requires further examination. Overall, the study supported the role of response inhibition in the DSM–IV ADHD combined type, but with key qualifications as to degree of specificity in reference both to comorbid problems and other executive functions.     

Integrity and Honesty Testing: What Do We Know? How Do We Use It?

This selective review of integrity and honesty testing addresses two primary questions: ‘What do we know about honesty testing?’ and ‘How do we use what we know?’ Up-to-date information about test reliability, validity, and construct definition from recent reviews of the research literature in the USA is presented and interpreted. Relationships to other selection devices and personality measures are discussed, as well as how integrity tests fit into a multiple assessment selection system.     

Sex Biases in Subject Selection: A Survey of Articles Published in American Medical Journals

This study discusses the results of a survey of 1,800 articles published in American medical journals from 1985--1996. The study finds 9% of these articles reported research that uses only male subjects to examine medical conditions that affect both sexes; the ratio of research on female to male conditions among these articles was greater than 5:1; but 76.5% of the articles reported research that includes both male and female subjects. The study also discusses evidence that sex biases against women (and men) are decreasing. This study also offers some possible psychological, institutional, medical, and economic explanations of the sex biases in medical research published in American journals, and discusses some policy implications of sex biases in medical research. The study concludes by urging others to conduct more empirical research on sex biases in medical research.     

大五人格理论与人事测评中的人格结构分析

本研究以工商企业人事选拔工作中参与实际测评的成人被试为研究对象,采用因素分析方法对中国成人人格的内隐结构进行了分析,并对所获取得主要特质因素与大五人格理论模型的关系和差别进行了探讨。     

Equivalence class establishment, expansion, and modification in preschool children

Preschool children were taught four two-choice match-to-sample conditional discriminations with 10 arbitrary visual stimuli. For 6 participants, 2 of the 10 stimuli served as the sample, or conditional, stimuli in all discriminations. For 5 additional participants, the same pair of stimuli served as the discriminative, or comparison, stimuli in all discriminations. Equivalence classes were established with more participants in the latter group, replicating prior research with participants with retardation. Four participants, in whom equivalence classes were established and who were available for further participation, were exposed to new conditional discriminations without trial-by-trial feedback and involving some novel and some familiar stimuli. Consistent conditional responding was observed, and tests for inclusion of the novel stimuli in the original classes showed class expansion. Training to reverse the unreinforced conditional performances produced a reversal of class membership in 3 of 4 participants, an outcome not consistent with other studies. The results are discussed with respect to the interaction of class structure and size.