Multilevel Modeling with Correlated Effects

When there exist omitted effects, measurement error, and/or simultaneity in multilevel models, explanatory variables may be correlated with random components, and standard estimation methods do not provide consistent estimates of model parameters. This paper introduces estimators that are consistent under such conditions. By employing generalized method of moments (GMM) estimation techniques in multilevel modeling, the authors present a series of estimators along a robust to efficient continuum. This continuum depends on the assumptions that the analyst makes regarding the extent of the correlated effects. It is shown that the GMM approach provides an overarching framework that encompasses well-known estimators such as fixed and random effects estimators and also provides more options. These GMM estimators can be expressed as instrumental variable (IV) estimators which enhances their interpretability. Moreover, by exploiting the hierarchical structure of the data, the current technique does not require additional variables unlike traditional IV methods. Further, statistical tests are developed to compare the different estimators. A simulation study examines the finite sample properties of the estimators and tests and confirms the theoretical order of the estimators with respect to their robustness and efficiency. It further shows that not only are regression coefficients biased, but variance components may be severely underestimated in the presence of correlated effects. Empirical standard errors are employed as they are less sensitive to correlated effects when compared to model-based standard errors. An example using student achievement data shows that GMM estimators can be effectively used in a search for the most efficient among unbiased estimators. This research was supported by the National Academy of Education/Spencer Foundation and the National Science Foundation, grant number SES-0436274. We thank the editor, associate editor, and referees for detailed feedback that helped improve the paper.     

Conditional relations with compound abstract stimuli using a go/no-go procedure

The aim of this study was to evaluate whether emergent conditional relations could be established with a go/no-go procedure using compound abstract stimuli. The procedure was conducted with 6 adult humans. During training, responses emitted in the presence of certain stimulus compounds (A1B1, A2B2, A3B3, B1C1, B2C2, and B3C3) were followed by reinforcing consequences (points); responses emitted in the presence of other compounds (A1B2, A1B3, A2B1, A2B3, A3B1, A3B2, B1C2, B1C3, B2C1, B2C3, B3C1 and B3C2) were not (i.e., extinction). During subsequent tests of emergent relations, new configurations (BA, CB, AC, and CA relations) were presented, formed by the recombination of training stimuli and structurally resembling tests usually employed in stimulus equivalence studies. Results showed that all 6 participants displayed immediate emergence of relations consistent with symmetry. Four participants exhibited emergent relations consistent with both transitivity and equivalence. These results indicate that a go/no-go procedure with compound stimuli can establish emergent conditional relations, thus providing a procedural alternative to the matching-to-sample procedures commonly used in studies of stimulus equivalence.     

Emergence of response‐response relations

The present study examined whether bidirectional response‐response relations could be established without direct reinforcement. In AB training for 12 undergraduates, higher rates of touches to a white circle on the monitor screen (A1) produced two stimuli (B1 and B2) on half of the trials, whereas lower rates (A2) produced the same effect on the other half. Choosing one of the two stimuli was reinforced according to the preceding responding (A1B1 and A2B2). In BC training, touching a stimulus (B1 or B2) produced three white circles lined up horizontally on the screen, after which one of two different response sequences to the circles (left‐center‐right, C1 or C2, and right‐center‐left, C2 or C1) were reinforced, depending on the stimulus presented (B1C1 and B2C2). After AB and BC relations were established, 11 of 12 participants showed the emergence of untrained relations (BA, CB, AC, and CA) throughout five test sessions, and the remaining participant showed it in the first four test sessions. These response‐response relations were replicated with five other undergraduates and different trained relations.     

词汇飞跃的本质:命名洞察力的实证研究

通过个案跟踪和控制实验探讨了词汇飞跃的本质。研究结果表明命名洞察力具有现实性,且词汇飞跃与命名洞察力的出现紧密相关,这在一定程度上解释了儿童词汇量快速增长的原因。命名洞察力的获得促进了词汇的快速增长,标志着儿童建立起词汇与指称之间的匹配关系。词汇飞跃之后,儿童很快建立起词语间的符号关系,并最终促进了儿童语法的发展。基于此实证结果,本研究还建构了诠释词汇飞跃本质的理论模型。     

Equivalence tests for comparing correlation and regression coefficients

Equivalence tests are an alternative to traditional difference‐based tests for demonstrating a lack of association between two variables. While there are several recent studies investigating equivalence tests for comparing means, little research has been conducted on equivalence methods for evaluating the equivalence or similarity of two correlation coefficients or two regression coefficients. The current project proposes novel tests for evaluating the equivalence of two regression or correlation coefficients derived from the two one‐sided tests (TOST) method (Schuirmann, 1987, J. Pharmacokinet. Biopharm, 15, 657) and an equivalence test by Anderson and Hauck (1983, Stat. Commun., 12, 2663). A simulation study was used to evaluate the performance of these tests and compare them with the common, yet inappropriate, method of assessing equivalence using non‐rejection of the null hypothesis in difference‐based tests. Results demonstrate that equivalence tests have more accurate probabilities of declaring equivalence than difference‐based tests. However, equivalence tests require large sample sizes to ensure adequate power. We recommend the Anderson–Hauck equivalence test over the TOST method for comparing correlation or regression coefficients.     

What kind of discipline is psychoanalysis?

Current controversies involving clinical, conceptual and empirical research shed light on how psychoanalysis confronts its nature and its future. Some relevant debates in which Wallerstein, Green, Hoffman, Eagle and Wolitzky, Safran, Stern, Blass and Carmeli, and Panksepp have participated are examined regarding the characteristics of their argumentation. Agreements and disagreements are explored to find ways that could have allowed the discussion to progress. Two foci are highlighted in these debates: (a) whether a clinical common ground exists in psychoanalysis and what kind of procedure could contribute to further clarification; (b) complementation of in‐clinical and extra‐clinical evidence. Both aspects are scrutinized: the possibility of complementing diverse methodologies, and the nature of the shared clinical evidence examined in clinical discussion groups such as those promoted by the IPA Clinical Observation Committee. The importance of triangulation and consilience is brought to bear regarding their contribution to the robustness of psychoanalysis. So as to strengthen a critical perspective that enhances the discipline's argumentative field, psychoanalysis should take into account arguments from different sources according to their specific merits. By doing this, psychoanalysis increases its relevance within the current interdisciplinary dialogue.     

On Murray Jackson's 1961 ‘Chair,couch and countertransference’

One of the problems facing psychoanalysts of all schools is that theory has evolved at a much faster pace than practice. Whereas there has been an explosion of theory, practice has remained, at least officially, static and unchanging. It is in this sense that Murray Jackson's 1961 paper is still relevant today. Despite the rise of the new relational and intersubjective paradigms, most psychoanalysts, and not a few Jungian analysts, still seem to feel that the couch is an essential component of the analytical setting and process. If the use of the couch is usually justified by the argument that it favours regression, facilitates analytical reverie and protects the patient from the influence of the analyst, over time many important psychoanalysts have come to challenge this position. Increasingly these analysts suggest that the use of the couch may actually be incompatible with the newer theoretical models. This contention is strengthened by some of the findings coming from the neurosciences and infant research. This underlines the necessity of empirical research to verify the clinical effectiveness of these different positions, couch or face‐to‐face, but it is exactly this type of research that is lacking.     

Moving to Learn: How Guiding the Hands Can Set the Stage for Learning

Previous work has found that guiding problem‐solvers' movements can have an immediate effect on their ability to solve a problem. Here we explore these processes in a learning paradigm. We ask whether guiding a learner's movements can have a delayed effect on learning, setting the stage for change that comes about only after instruction. Children were taught movements that were either relevant or irrelevant to solving mathematical equivalence problems and were told to produce the movements on a series of problems before they received instruction in mathematical equivalence. Children in the relevant movement condition improved after instruction significantly more than children in the irrelevant movement condition, despite the fact that the children showed no improvement in their understanding of mathematical equivalence on a ratings task or on a paper‐and‐pencil test taken immediately after the movements but before instruction. Movements of the body can thus be used to sow the seeds of conceptual change. But those seeds do not necessarily come to fruition until after the learner has received explicit instruction in the concept, suggesting a “sleeper effect” of gesture on learning.