Testing for DIF in a model with single peaked item characteristic curves: The parella model

The PARELLA model is a probabilistic parallelogram model that can be used for the measurement of latent attitudes or latent preferences. The data analyzed are the dichotomous responses of persons to items, with a one (zero) indicating agreement (disagreement) with the content of the item. The model provides a unidimensional representation of persons and items. The response probabilities are a function of the distance between person and item: the smaller the distance, the larger the probability that a person will agree with the content of the item. This paper discusses how the approach to differential item functioning presented by Thissen, Steinberg, and Wainer can be implemented for the PARELLA model. Requests for the PARELLA software should be sent to Iec Progamma PO Box 841, 9700 AV Groningen, The Netherlands.     

The intrinsic difficulty of recursive functions

This paper deals with a philosophical question that arises within the theory of computational complexity: how to understand the notion of INTRINSIC complexity or difficulty, as opposed to notions of difficulty that depend on the particular computational model used. The paper uses ideas from Blum's abstract approach to complexity theory to develop an extensional approach to this question. Among other things, it shows how such an approach gives detailed confirmation of the view that subrecursive hierarchies tend to rank functions in terms of their intrinsic, and not just their model-dependent, difficulty, and it shows how the approach allows us to model the idea that intrinsic difficulty is a fuzzy concept. Jan Zygmunt     

Optimum examinee samples for item parameter estimation in item response theory: A multi-objective programming approach

This paper proposes a multi-objective programming method for determining samples of examinees needed for estimating the parameters of a group of items. In the numerical experiments, optimum samples are compared to uniformly and normally distributed samples. The results show that the samples usually recommended in the literature are well suited for estimating the difficulty parameters. Furthermore, they are also adequate for estimating the discrimination parameters in the three-parameter model, butnot for the guessing parameters.     

Acceleration model in the heterogeneous case of the general graded response model

A new model, called acceleration model, is proposed in the framework of the heterogenous case of the graded response model, based on processing functions defined for a finite or enumerable number of steps. The model is expected to be useful in cognitive assessment, as well as in more traditional areas of application of latent trait models. Criteria for evaluating models are proposed, and soundness and robustness of the acceleration model are discussed. Graded response models based on individual choice behavior are also discussed, and criticisms on model selection in terms of fitnesses of models to the data are also given.This research was supported by the Office of Naval Research (N00014-90-J-1456).     

Projection of a binary criterion into a model of hierarchical classes

A formal analysis is made of how to project an attribute criterion into the hierarchical classes model for object by attribute data proposed by De Boeck and Rosenberg. The projection is conceptualized as the prediction of the attribute criterion by means of a logical rule defined on the basis of attribute combinations from the model. Eliminative and constructive strategies are proposed to find logical rules with maximal predictive power and minimal formula complexity. Logical analyses of a real data set are reported and compared with a logistic regression to demonstrate the usefulness of the logical strategies, and to show the complementarity of logical and probabilistic approaches.The first suthor is Senior Research Assistant of the National Fund for Scientific Research (Belgium). We would like to thank the Editor, the reviewers, Seymour Rosenberg, and Luc Delbeke for their helpful comments on earlier drafts of this article.     

Specifying optimum examinees for item parameter estimation in item response theory

Information functions are used to find the optimum ability levels and maximum contributions to information for estimating item parameters in three commonly used logistic item response models. For the three and two parameter logistic models, examinees who contribute maximally to the estimation of item difficulty contribute little to the estimation of item discrimination. This suggests that in applications that depend heavily upon the veracity of individual item parameter estimates (e.g. adaptive testing or text construction), better item calibration results may be obtained (for fixed sample sizes) from examinee calibration samples in which ability is widely dispersed.This work was supported by Contract No. N00014-83-C-0457, project designation NR 150-520, from Cognitive Science Program, Cognitive and Neural Sciences Division, Office of Naval Research and Educational Testing Service through the Program Research Planning Council. Reproduction in whole or in part is permitted for any purpose of the United States Government. The author wishes to acknowledge the invaluable assistance of Maxine B. Kingston in carrying out this study, and to thank Charles Lewis for his many insightful comments on earlier drafts of this paper.     

Quantification of rats' behavior during reinforcement periods

What is treated as a single unit of reinforcement often involves what could be called a reinforcement period during which two or more acts of ingestion may occur, and each of these may have associated with it a series of responses, some reflexive, some learned, that lead up to ingestion. Food-tray presentation to a pigeon is an example of such a “reinforcement period.” In order to quantify this behavior, a continuous-reinforcement schedule was used as the reinforcement period and was chained to a fixed-ratio schedule. Both fixed-ratio size and reinforcement-period duration were manipulated. Rats were used as subjects, food as reinforcement, and a lever press as the operant. Major findings included (a) a rapid decline in response rates during the first 15 to 20 seconds of the reinforcement periods, and (b) a strong positive relationship between these response rates and the size of the fixed ratio. Also revealed was a short scallop not normally found in fixed-ratio response patterns, whose length was a function of fixed-ratio size and reinforcement-period duration. It is suggested that rapidly fluctuating excitatory processes can account for many of these findings and that such processes are functionally significant in terms of behavioral compensation.     

Effects of relational uncertainty in heightening national identification and reactive approach motivation of Japanese

This study investigated whether relational uncertainty poses uncertainty threat, which causes compensatory behaviours among Japanese. We hypothesised that Japanese, as collectivists, would perceive relational uncertainty to pose uncertainty threat. In two experiments, we manipulated relational uncertainty, and confirmed that participants exhibited compensatory reactions to reduce aversive feelings due to it. In Study 1, we conducted direct comparison between relational uncertainty, independent self‐uncertainty and control conditions. The results revealed that participants who were instructed to imagine events pertaining to relational uncertainty heightened national identification as compensation than did participants in the control condition, but independent self‐uncertainty did not provoke such effects. In Study 2, we again manipulated relational uncertainty; however, we also manipulated participants' individualism‐collectivism cultural orientation through priming, and the analyses yielded a significant interaction effect between these variables. Relational uncertainty evoked reactive approach motivation, a cause for compensatory behaviours, among participants primed with collectivism, but not for individualism. It was concluded that the effect of uncertainty on compensatory behaviour is influenced by cultural priming, and that relational uncertainty is important to Japanese.     

A performance comparison between the two language versions of the Affective Go/No‐Go test: A randomised crossover study

The purpose of this randomised crossover study is to validate the Korean version of the Affective Go/No‐go (AGN) test. The Korean words for the AGN test were selected after careful evaluation of emotional valences, word length and frequency. Fifteen Korean advanced learners of English were administered both Korean and English versions, yielding 30 data points. The performance of both language versions was compared for each of the AGN test parameters (response latency, commission error and omission error) using two‐way analysis of variance (ANOVA). Intraclass correlation coefficients (ICCs) were estimated to evaluate associations between the two versions. The ICCs were high for response latencies of all valences and commission errors of positive and neutral words, but not for that of negative words and omission errors of all valences. A similar pattern of test results, as revealed by the high ICCs and non‐significant interaction effects between language and word valence, suggests that the psychometric properties of the AGN test may be comparable over different language versions.     

Risk compensation theory and bicycle helmets – Results from an experiment of cycling speed and short-term effects of habituation

It has been suggested that the safety benefits of bicycle helmets are limited by risk compensation. The current study contributes to explaining whether the potential safety effects of bicycle helmets are reduced by cyclists’ tendency to cycle faster when wearing them (as a result of risk compensation), and if this potential reduction can be associated with a change in perceived risk. A previous study (Fyhri & Phillips, 2013) showed that non-routine helmet users did not increase their speed immediately after being given a helmet to wear, while routine helmet users cycled more slowly. The current study tests whether the previously found reduction in speed in response to helmet removal – as an indirect indicator of risk compensation – could be established in non-routine helmet users, after a period of habituation while cycling with a helmet.We did this by conducting a randomized crossover trial, in which we used GPS-derived speed calculations and self-reported risk perception. To test the effect of habituation, we used a design where each participant took part in two rounds with a break between and each round having two trips. We collected the data in June 2015. Non-routine helmet users (N = 31) were recruited in the field (along cycle routes in Oslo), and through a sample drawn from the Falck National register of bicycle owners. In the first phase of the study, all participants were asked to complete a test route (2.4 km downhill) with and without a helmet. In the second phase of the experiment, conducted after 1.5–2 h, the same participants again completed the test route with and without a helmet. In the time between the first and second phases of the experiment, all participants were given helmets, and told to use them on a predefined bicycle route.Habituation to the helmet between the first and second phases of the experiment did not produce any decrease (with helmet removal) in speed, on top of the habituation that occurred while cycling down the hill (the order effect). Mean speed difference for cycling with/without a helmet before the break was −0.76 km/h, after the break this difference was 0.32 km/h; 95% CIs [−0.5, 2.9] and [−0.9, 1.5]. We argue that risk compensation is an unlikely effect of using a bicycle helmet, and probably cannot explain any adverse effects related to helmet legislation.