Behaviormetrika

Online ISSN: 1349-6964
Print ISSN: 0385-7417
Publications
"Multidimensional scaling technique was applied to the Brazilian migration matrix in order to identify the functional regions of the nation, using the 1970 census data for males. The results obtained from three types of proximity matrices confirmed the key roles of Minas Gerais and Mato Grosso as the core states in addition to Rio de Janeiro and Sao Paulo. Substantial circulation of migrants among the core states runs counter to the popular notion about the Northeastern states as the major source of labor in Rio de Janeiro and Sao Paulo. The Northeast [was] found to consist of three groups of states: Maranhao, Piaui and Ceara showed a stable clustering, but peripherally located, in all configurations; and, the other two groups failed to form persistent regions across configurations. Though limited in scope, the observed attraction of Espirito Santo and Goias violated the widely held negative effect of distance and deserves further investigation."
 
Coordinates of 17 objects in two-dimensional space
δ-monotone admissibilities of 8 AHCAs: Determined after analysis
This paper discusses the admissibility of agglomerative hierarchical clustering algorithms with respect to space distortion and monotonicity, as defined by Yadohisa et al. and Batagelj, respectively. Several admissibilities and their properties are given for selecting a clustering algorithm. Necessary and sufficient conditions for an updating formula, as introduced by Lance and Williams, are provided for the proposed admissibility criteria. A detailed explanation of the admissibility of eight popular algorithms is also given.
 
Studies of sociopolitical attitudinal shifts toward increased conservatism have failed to adequately distinguish among period, age and cohort effects. Using a modified Bayesian cohort procedure and data from the General Social Survey, 1972–1990, we examine these effects and test 3 hypotheses: (1) period effects will run in a conservative direction for those variables related to morality and crime, and in a liberal direction for those variables related to women's rights and civil liberties; (2) Age effects will run in a conservative direction for those variables related to women's rights, civil liberties and morality; and (3) Post-baby boom cohort effects will run in a conservative direction for those variables related to morality and civil liberties. Results show that period shifts have occurred in the conservative direction for variables concerning moral issues and crime, but in the liberal direction for civil liberties and women's rights. Increased age is associated with increased conservatism with regard to women's rights and homosexuality, but otherwise, does not lead to increased conservatism. Cohort effects show that more recent cohorts are increasingly conservative in their attitudes toward drug use and homosexuality, but are otherwise relatively liberal. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
One of the widely acknowledged drawbacks of flexible statistical models is that they are often extremely difficult to interpret. However, if flexible models are constrained to be additive they are much easier to interpret, as each input can be considered independently. The problem with additive models is that they cannot provide an accurate model if the phenomenon being modeled is not additive. This paper proposes that a tradeoff between accuracy and additivity can be implemented easily in a particular type of flexible model: a Gaussian process model. One can build a series of Gaussian process models which begin with the completely flexible and are constrained to be more and more additive, and thus interpretable. Observations of how the test error and importance of interactions change as the model becomes more additivegive insightinto the importance and nature of interactions. Models in the series can also be interpreted graphically with a technique for visualizing the effects...
 
Examines geometrical structures of some non-distance models for asymmetric multidimensional scaling (MDS) in error-free data measured at a ratio level, and presents unified geometrical interpretation of these models. These include Canonical Analysis of SKew symmetry (CASK), DEDICOM, GIPSCAL, and Hermitian Canonical Model. It is shown that these models, except for CASK and other possible models for square asymmetric proximity data matrix, are expressible in terms of finite-dimensional complex Hilbert space under some general condition, and that differences in form of these models depend only on the bases chosen. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Suggests that identifying the elements that make up an author's characteristic writing style is one of the keys to determining the authorship or authenticity of a literary composition. The characters that commas follow are analyzed in works by 4 Japanese authors. The placement of commas differs from writer to writer and may be considered one of the features that makes up an individual literary style, thus providing valuable information for verifying authenticity and speculating on questions of authorship. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
The normal theory maximum likelihood and asymptotically distribution free methods are commonly used in covariance structure practice. When the number of observed variables is too large, neither method may give reliable inference due to bad condition numbers or unstable solutions. The main existing solution to the problem of high dimension is to build a model based on marginal variables. This practice is inefficient because the omitted variables may still contain valuable information regarding the structural model. In this paper, we propose a simple method of averaging proper variables which have similar factor structures in a confirmatory factor model. The effects of averaging variables on estimators and tests are investigated. Conditions on the relative errors of the measured variables are given that verify when a model based on averaged variables can give better estimators and tests than one based on omitted variables. Our method is compared to the method of variable selection based on mean square error of predicted factor scores. Some aspects related to averaging, such as improving the normality of observed variables, are also discussed. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Cultural link analysis is used for the comparative study of the social attitudes of people in different cultures (France, Germany, UK, US born on mainland or in Hawaii, and Japanese born in Japan or Hawaii), using 5 nations' surveys. Cultural link analysis is a statistical method relatively free from cultural and ideological biases that provides information on the sample, the item content, and stability and change over time. Analyses of various sets of questions (trustfulness and work; money, work, and national goals; and Aesop's fables and related social attitudes) and of various scales (on every nation; by national and concept scale; by nation, scales, and other questions) highlighted the value systems of various nations and showed similarity and dissimilarity, and universality and particularity. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
89 undergraduates were required to determine a value of the ambiguous parameter in the "problem of 3 prisoners" (S. Shimojo and S. Ichikawa; see record 1989-38856-001) and then solve the problem. This premanipulation did not improve performance: most Ss could not give the answer in accordance with the parameter they set for themselves. Many Ss held a crucial fallacy on the relation between prior and posterior probabilities. The difficulty of the problem appears to be with Ss' erroneous beliefs about the nature of probability rather than setting a value of the parameter. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
A mathematical analysis of magical numbers on human memory span shows that if memory is hierarchically organized the optimal chunk size of short-term memory (STM) may be 7.34…±1.74… for an exhaustive search strategy and 3.59…±0.49… for a self-terminating search strategy. Further, it illustrates how the optimal form of memory may change in a trade-off between memory organization cost and searching cost (or time). Comparison of the theoretical estimates with previous empirical estimates demonstrates that limitations of human STM capacity may be related to the efficiency of design. It is also demonstrated that the coding system of DNA may be interpreted as a means of efficient information retrieval as in human STM. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Used computer graphic representation to measure individual parts of the characters and to define many variables for each character. Since variables important in the discriminant analysis varied among sample characters, the stepwise discriminant analysis was useful. (20 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Discusses the use of penalty marks in competitive multiple-choice examinations and evaluates 2 systems—the gap coefficient and the inversion rate—of crediting points for a 5-choice item under certain assumptions on the distribution of true scores. Although some assumptions are arbitrary, it is shown that, in some cases, penalty marks are able to reduce the possibility of less knowledgeable examinees attaining higher scores than more knowledgeable examinees. (6 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Described 2 cohorts of longitudinal data concerned with early predictors of inhibited and uninhibited temperamental types of infants. There were 93 Ss in Cohort 1 and 76 Ss in Cohort 2. Ss were tested at 4 mo and at 14 mo of age to determine their reactions to changes in the intensity and variety of sensory stimuli and to assess fear of unfamiliar events, respectively. A finite mixture model was used to analyze the data, and a measure of predictive efficacy was described for comparing the mixture model with competing models (linear regression analysis). The mixture model performed mildly better than the linear regression model with respect to this measure of fit to the data. However, the primary advantage of the mixture model relative to competing approaches was that it could be easily used to address improvements and corrections to the theory behind the research, and to suggest extensions of the research. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Introduces a probabilistic model that is capable of diagnosing and classifying cognitive errors in a general problem-solving domain. Item response theory is used to handle the variability of response errors. The model is illustrated by the classification of 595 junior high school students' responses on a 38-item fraction addition test into 34 groups of misconceptions that were predetermined by an error analysis previously done and validated with the error diagnostic program written by a typical formal logic approach. An important characteristic of the model is the possibility of controlling classifications of performances by selecting meaningful sets of ellipses, representing erroneous rules in a procedural domain, sources of misconception producing erroneous rules of operation, or different achievement levels. The model is a generalized version of mastery or criterion testing and thus is applicable to computer-aided instruction. (20 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Presents results from analyses of 3 items in the National Character Survey, a questionnaire study providing time-series data covering a 25-yr period. Surveys were carried out every 5 yrs, beginning in 1953, to examine the changes in attitudes that have taken place in post-war Japan. Analyses suggest that shifts in attitudes among Japanese adults were not due solely to the process of population replacement. Even past their formative years, individuals changed their views, and often quite rapidly. Shifts in attitudes in different areas were not always synchronous. The impact of social, economic, and political developments was not uniform across different attitudinal areas. There were important attitudinal cleavages, the most notable one based on sex. In all 3 items, significant sex-differences were found in responses. (25 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Reviews recent theoretical and methodological developments in multidimensional scaling (MDS) and its related methods in Japan and introduces relevant Japanese work. The term MDS is used in 2 senses. In the narrow sense, it is a technical term that is divided into metric MDS and nonmetric MDS. The wider sense includes many variations in multidimensional data analysis. Both approaches are taken into consideration in clarifying the main methodological points of the various works. (93 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Any exploratory factor analysis model requires at least three indicators (observed variables) for each common factor to ensure model identifiability. If one makes exploratory factor analysis for a data set in which one of common factors would have only two indicators in its population, one would encounter difficulties such as improper solutions and nonconvergence of iterative process in calculating estimates. In this paper, we first develop conditions for partial identifiability of the remaining factor loadings except for a factor loading vector which relates to a common factor with only two indicators. Two models for analyzing such data sets are then proposed with the help of confirmatory factor analysis and covariance structure analysis. The first model is an exploratory factor analysis model that permits correlation between unique factors; the second model is a kind of confirmatory factor model with equal factor loadings. Two real data sets are analyzed to illustrate usefulness of t...
 
Features gathered from the observation of a phenomenon are not all equally informative: some of them may be noisy, correlated or irrelevant. Feature selection aims at selecting a feature set that is relevant for a given task. This problem is complex and remains an important issue in many domains. In the field of neural networks, feature selection has been studied for the last ten years and classical as well as original methods have been employed. This paper is a review of neural network approaches to feature selection. We first briefly introduce baseline statistical methods used in regression and classification. We then describe families of methods which have been developed specifically for neural networks. Representative methods are then compared on different test problems. Keywords Feature Selection, Subset selection, Variable Sensitivity, Sequential Search Sélection de Variables et Réseaux de Neurones Philippe LERAY et Patrick GALLINARI Résumé Les données collectées lors de l'obse...
 
Proposed is GLLL2, a hybrid architecture of a global and a local learning module, which learns default and exceptional knowledge respectively from noisy examples. The global learning module, which is a feedforward neural network, captures global trends gradually, while the local learning module stores local exceptions quickly. The latter module distinguishes noise from exceptions, and learns only exceptions, which ability makes GLLL2 noise-tolerant. The results of experiments show the process in which training examples are formed into default and exceptional knowledge, and demonstrate that the predictive accuracy, the space efficiency, and the training efficiency of GLLL2 is higher than those of each individual module. Introduction In cognitive science, whether learners induce rules or remember exemplars has been at issue recently. In particular, learning processes of quasi-regular tasks that involve both regularities and exceptions has attracted considerable attention. We developed G...
 
Proposes a graphical rank test using the rank graph, a descriptive graph in which the average rank and the degree of concordance of ranks assigned to each item are represented by an item vector. Under the assumption of random rankings, the distribution of the final point of an item vector can be asymptotically approximated by a 2-dimensional normal distribution. In the case of large sample, the null hypothesis for each item can be tested visually by the rank graph with a critical ellipse. (17 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Proposes a model of extended lateral inhibition that involves the size of the receptive field as a parameter. Variations in the size of the receptive field indicate functional differences between central and peripheral vision. The model is applied to the study of differences in brightness sensitivity between the fovea and the peripheral area, as observed in the Hermann grid illusion. The model is also numerically simulated on a computer. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Starting from data available about (anonymous) visitors of a website, possibilities of extracting information as well as enriching pure browsing patterns with the help of business measures (e-metrics) and detecting possibly negative effects of web robots are sketched as prerequisites for interpretation purposes of the navigational behavior of internet users. Problems with the reconstruction of user navigation paths are explained and different heuristics for path completion are discussed. Additionally, several kinds of features of user navigation paths (e.g., sets, sequences, path fragments) have to be mentioned to prepare for an adequate theoretical background concerning recommender systems that can be used for tasks as different as site personalization, cross-/ up-selling, and navigation assistance. A vocabulary to describe different kinds of recommender systems and generic quality measures for system evaluation are formulated. Then, specific recommender systems, especially systems based on frequent path features, are defined and evaluated in a final experiment. In an outlook directions for future research on recommender systems are given. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Presents a multidimensional unfolding method for rank order data (MDU-RANK), incorporating a goodness of fit criterion that is directly linked to the response model of ranking behavior so that it can indicate the degree to which the response model fits the data. Owing to this criterion, the MDU-RANK method never fails to yield stable solutions free from degeneration. Unlike MDU-RANK, conventional nonmetric unfolding methods are frequently bothered by degeneracies caused by minimizing the so-called stress measure at the expense of order correspondences between the data and the model. The MDU-RANK method is applied to 2 sets of real data for illustrative purposes. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Examined the perception of profiles as solid objects in motion in 2 experiments. Exp 1, involving 9 undergraduates, explored the effects of the motion of points of correspondence when the viewing direction was perpendicular to the axis of rotation. Results support the hypothesis that the motion of points of correspondence provides a salient cue to motion in depth of a solid object. Exp 2, involving 12 undergraduates, investigated how the perception of profiles changed when the viewing direction was no longer perpendicular to the axis of rotation. Even without the simple trajectories, Ss reported profiles as the motion of a rigid object in depth. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
 
Plot of subject estimates for the 2-PM dataset; MAP (horizontal) versus HA (vertical).  
Plot of discrimination parameter estimates for the 2-PM dataset; MML (horizontal) versus HA (vertical).  
Plot of difficulty parameter estimates for the 2-PM dataset; MML (horizontal) versus HA (vertical).
Plot of discrimination parameter estimates for the Rasch dataset; MML (horizontal) versus HA (vertical).
Plot of difficulty parameter estimates for the Rasch dataset; MML (horizontal) versus HA (vertical).  
Two test theoretical approaches to item analysis are compared, an approach based on homogeneity analysis and one based on item response theory. The literature on the relationship between the two approaches is briefly reviewed. The paper contains a contribution to the relationship between the two approaches for the case that the scores are dichotomous and a single latent variable is assumed to underlie the data. A loss function is proposed for modelling item response functions with two parameters, one for discrimination and one for difficulty. It turns out that the loss of the proposed loss function is related to loss of homogeneity. Demonstrations with simulated data are used to evaluate the proposed method.
 
Three-way interaction effect of relative class size, number of variables and clustering technique on cluster recovery
Conditional probabilities of Simulation Study I
Means and standard deviations (in parentheses) of the adjusted Rand index in Simulation Study I
Test statistics of effects on cluster recovery in Simulation Study I effect SS (type III) df F partial η 2
Proportions of used solution strategies
The problem considered in the present paper is how to cluster data of nominal measurement level, where the categories of the variables are equivalent (the variables are replications of each other). One suitable technique to obtain such a clustering is latent class analysis (LCA) with equality restrictions on the conditional probabilities. As an alternative, a less well known technique is introduced: GROUPALS. This is an algorithm for the simultaneous scaling (by multiple correspondence analysis) and clustering of categorical variables. Equality restrictions on the category quantifications were incorporated in the algorithm, to account for equivalent categories. In two simulation studies, the clustering performance was assessed by measuring the recovery of true cluster membership of the individuals. The effect of several systematically varied data features was studied. Restricted LCA obtained good to excellent cluster recovery results. Restricted GROUPALS approximated this optimal performance reasonably well, except when underlying classes were very different in size.
 
In the paper proposed we will make use of the gradient flow approach to consider a generalization of the well-known oblique Procrustes rotation problem, involving oblique simple structure rotation of both the core and component matrices resulting from three-mode factor analysis. The standard oblique Procrustes rotations to specified factor-structure and factor-pattern follow as special cases. The approach adopted leads to globally convergent algorithm and includes solving of initial value problem for certain matrix ordinary differential equation. Necessary conditions are established for the solution of the problem. The same approach is extended easily to the weighted oblique Procrustes rotation. Finally, some simulated numerical results are given and commented.
 
In the present paper the ORTHOMAX rotation problem is reconsidered. It is shown that its solution can be presented as a steepest ascent flow on the manifold of orthogonal matrices. A matrix formulation of the ORTHOMAX problem is given as an initial value problem for matrix differential equation of first order. The solution can be found by any available ODE numerical integrator. Thus the paper proposes a convergent method for direct matrix solution of the ORTHOMAX problem. The well-known first order necessary condition for the VARIMAX maximizer is reestablished for the ORTHOMAX case without using Lagrange multipliers. Additionally new second order optimality conditions are derived and as a consequence an explicit second order necessary condition for further classification of the ORTHOMAX maximizer is obtained.
 
It was proposed that people could evaluate the other person's evaluation scheme in terms of the similarity (or dissimilarity) of the other person's evaluation scheme to their own schemes. The metric for the dissimilarity of evaluation schemes was proposed to be the Minkowski-p metric for the utilities evaluated under different schemes. This model provides a basis for allocating importance to utility functions, where importance is to be interpreted as the similarity of the integrated scheme to the schemes represented by the utility functions. An experiment was done to investigate whether people's intuitive judgments for the dissimilarity of evaluation schemes could be described by the Minkowski-p metric; the results generally supported the model and also suggested that the metric would be the “city-block” metric, i.e., p=1 in the Minkowski-p metric.
 
The present study investigated the consequences of ignoring a nested data structure on the Rasch/one parameter item response theory model. Although most large-scale educational assessment data do exhibit a nested data structure, current practice often ignores such data structure and applies the standard Rasch/IRT models to conduct measurement analyses. We hypothesized that this practice would produce negative consequences on the item parameter estimates. Using simulation, we investigated this hypothesis by comparing the results from an incorrectly specified two level model which ignored the nested data structure to those from a correctly specified three-level hierarchical generalized linear model. Use of the incorrect two-level model did, in fact, result in negative consequences in estimating the standard errors, although the point estimates were unbiased and identical to the ones from the three-level analysis. A real data set from the IEA Civic education study in 1999 was used to illustrate the simulation results.
 
The Institute of Statistical Mathematics has been conducting a longitudinal survey on Japanese national character since 1953. From 1971, this survey was extended to include cross-national comparative surveys and people of Japanese ancestry in Hawaii, the West Coast of the United States, and Brazil. The cross-national survey primarily focuses on comparing social values, ways of thinking and feeling, and other relevant characteristics of people from various nations. This study investigates conditions under which meaningful cross-national comparability of social survey data is guaranteed, despite differences in languages and statistical sampling methods. Over the past 14 years, focusing on Asian countries, we have carried out the East Asia Values Survey (2002-2005), the Pacific Rim Values Survey (2004-2009), and the Asia-Pacific Values Survey (2010-2014). In this introductory paper to the special issue, we discuss the development of our research paradigm, termed cultural manifold analysis (CULMAN), and provide an overview of our past surveys. We conclude with comments on our future research.
 
The main purpose of this study is to investigate influence of nonresponse in the “Interview Survey for Stratification and Social Psychology in 2010” (SSP-I2010 Survey). Now, social stratification is one of main research themes in the study of Japanese society, and the SSP-I2010 Survey provides basic data to study social stratification and people’s views on economic inequality in Japan. From a target sample of 3,500, approximately half (1,737) did not respond in the survey, thus nonresponse bias is a serious concern. From a survey methodological viewpoint, studies applying methods for dealing with nonresponse to Japanese surveys are few. Therefore many empirical studies with nonresponse bias adjustment are needed to understand influence of nonresponse in Japanese surveys. In an attempt to reduce the nonresponse bias in the SSP-I2010 Survey, we used two bias adjustment methods using information on both survey locations and individuals as auxiliary variables. The effectiveness of the bias adjustment methods was evaluated by a simulation and several items of the SSP-I2010 Survey where the values of population proportions are known. In this study, stratum identification was relatively insensitive to bias adjustment. On the other hand, the estimates of the proportion of people who accept the economic inequality increased by bias adjustment.
 
This article introduces the articles in Behaviormetrika Vol. 44, No. 2, 2017.
 
Top-cited authors
Yoshio Takane
  • University of Victoria
Heungsun Hwang
  • McGill University
Thomas L. Saaty
  • University of Pittsburgh
Philippe Leray
  • University of Nantes
Ryozo Yoshino
  • Doshisha University