The EM algorithm for latent class analysis with equality constraints

The EM algorithm is a popular iterative method for estimating parameters in the latent class model where at each step the unknown parameters can be estimated simply as weighted sums of some latent proportions. The algorithm may also be used when some parameters are constrained to equal given constants or each other. It is shown that in the general case with equality constraints, the EM algorithm is not simple to apply because a nonlinear equation has to be solved. This problem arises, mainly, when equality constrints are defined over probabilities indifferent combinations of variables and latent classes. A simple condition is given in which, although probabilities in different variable-latent class combinations are constrained to be equal, the EM algorithm is still simple to apply.The authors are grateful to the Editor and the anonymous reviewers for their helpful comments on an earlier draft of this paper. C. C. Clogg and R. Luijkx are also acknowledged for verifying our results with their computer programs MLLSA and LCAG, respectively.     

题材个体化和问题重述两策略的效果研究

探讨题材个体化和问题重述作为两种解题策略在解决加、减法应用题中的有效性,采用了４种策略类型×３种数学能力水平×２种问题题材×２种题材情境与解题操作关系的四因素混合设计。结论如下：学生基本上能够独立进行题材个体化和问题重述;在本实验条件下,水平高的学生,运用不同策略没有差异;水平中等的学生运用双策略、水平低的学生运用双策略以及题材个体化策略均产生了积极效果;题材个体化策略对于解决不熟悉题材的应用题将产生积极效果,运用以上两策略解决问题情境与解题操作相矛盾的问题没有帮助。除了应用题陈述图式以外,本实验证明还存在着应用题结构图式;可把题材个体化和问题重述作为解题策略教给学生运用。     

Perceived social context and risk preference: A re-examination of framing effects in a life-death decision problem

This study examines the effects of perceived group context on subjects risk attitudes and their sensitivity to the framing of choice outcomes in a ‘life-death’ decision problem. It seeks to uncover the psychological mechanisms underlying decision-making biases by systematically manipulating the decision context in which the ‘life-death’ problem was described. The study revealed that subjects risk preferences varied as a function of the experimental manipulations. Previously observed reversals in preferences (framing effects) appeared in large-group contexts and disappeared in small-group and family contexts. When considering the fate of small groups, subjects unambiguously favored the probabilistic outcome, no matter how the ‘life-death’ decision problem was framed. The empirical data obtained from the present study suggest that human choice patterns are behaviorally distinguishable across large-group, small-group, and family social contexts.     

Some Thoughts on Language Use in Therapy

Drawing on the work of Ludwig Wittgenstein and his own experience as a therapist and educator/trainer, the author describes some of the issues involved in helping therapists to find their way out of various muddles and mazes that are deeply embedded in language.     

Why Logic Doesn‘t Matter in the (Philosophical) Study of Argumentation

Philosophically, the study of argumentation is important because it holds out the prospect of an interpretation of rationality. For this we need to identify a transcendent perspective on the argumentative interaction. We need a normative theory of argumentation that provides an answer to the question: should the hearer accept the argument of the speaker. In this article I argue that formal logic implies a notion of transcendence that is not suitable for the study of argumentation, because, from a logical point of view, argumentation disappears from sight. We should therefore not expect formal logic to provide an interesting interpretation of the rationality intrinsic in argument and discussion.     

The effects of nodality on the formation of equivalence classes

A four-member equivalence class (A → B → C → D) can be formed by training AB, BC, and CD. The nodal stimuli, B and C, mediate all of the derivative (transitive and equivalence) relations in the class. The derivative relations AC, CA, BD, and DB are separated by one node, whereas AD and DA are separated by two nodes. How do the number of nodes that separate the stimuli in a derivative relation influence the induction of stimulus control exerted by that relation? Seven college students learned two four-member classes made up of nonsense syllables. After training, all derivative relations were presented repeatedly without informative feedback. Stimulus control exerted by each derivative relation was assessed concurrently. For the 7 subjects, control exerted by the derivative relations increased gradually with repeated presentations. With 6 of the 7 subjects, the one-node relations exerted more control than the two-node relations during the process. However, the disparity between the one- and two-node relations decreased with repeated presentations. Eventually, all derivative relations exerted complete control. The control exerted by derivative relations during induction was inversely related to the number of nodes separating the terms in the derivative relations. These results demonstrate that nodal distance is a determinant of the relatedness of stimuli in equivalence classes. The findings are discussed in terms of remote association, semantic memory networks, and the study of transitive inference.     

Intentional explanations and radical theories of education

In this paper I argue that Marxist studies of schools have overlooked the power of intentional explanations to explain schooling practices and policies. This oversight is at least in part due to many radical analyses failing to distinguish between explaining the acquisition and persistence of beliefs and determining the social consequences that follow from acting on beliefs. I further contend that radical researchers examining schooling practices must develop a more rigorous and refined conception of capitalist class interests.     

The Dutch Identity: A new tool for the study of item response models

The Dutch Identity is a useful way to reexpress the basic equations of item response models that relate the manifest probabilities to the item response functions (IRFs) and the latent trait distribution. The identity may be exploited in several ways. For example: (a) to suggest how item response models behave for large numbers of items—they are approximate submodels of second-order loglinear models for 2 ^J tables; (b) to suggest new ways to assess the dimensionality of the latent trait—principle components analysis of matrices composed of second-order interactions from loglinear models; (c) to give insight into the structure of latent class models; and (d) to illuminate the problem of identifying the IRFs and the latent trait distribution from sample data.This research was supported in part by contract number N00014-87-K-0730 from the Cognitive Science Program of the Office of Naval Research. I realized the usefulness of the identity in Theorem 1 while lecturing in the Netherlands during October, 1986. Because this was in no small part due to the stimulating psychometric atmosphere there, I call the result the Dutch Identity.     

On the uses of language in working and idealized logic

The interpretation of discourse covers a continuum with two extremes: on the one hand, a text considered as an ideal, distant object, and on the other hand, a conversation regarded as a real, present event. On the basis of a distinction between relatively context-invariant propositions and relatively context-dependent statements, it is argued that statements in conversational discourse are easier to interpret than statements in texts, whereas only propositions in symbolic logic can be interpreted with exactitude. In the same way, the interpretation of dialogical arguments proceeds more easily than the interpretation of arguments in texts. While dialogical argumentation requires a dialectical approach, textual argumentation necessitates an imaginative reconstruction of the argument. From this it can be concluded that for different sorts of argumentative discourse diverse sorts of interpretative activities have to be used.     

Latent class models for nonmonotone dichotomous items

Starting from perfectly discriminating nonmonotone dichotomous items, a class of probabilistic models with or without response errors and with or without intrinsically unscalable respondents is described. All these models can be understood as simply restricted latent class analysis. Thus, the estimation and identifiability of the parameters (class sizes and item latent probabilities) as well as the chi-squared goodness-of-fit tests (Pearson and likelihood-ratio) are free of the problems. The applicability of the proposed variants of latent class models is demonstrated on real attitudinal data.This research was supported by the Kulturamt der Stadt Wien, Magistratsabteilung 7.The author wishes to thank the editor, Ivo W. Molenaar, as well as Clifford C. Clogg and the anonymous reviewers for their valuable comments on the earlier drafts of this paper.