Conditional logistic regression with sandwich estimators: application to a meta-analysis

Motivated by a meta-analysis of animal experiments on the effect of dietary fat and total caloric intake on mammary tumorigenesis, we explore the use of sandwich estimators of variance with conditional logistic regression. Classical conditional logistic regression assumes that the parameters are fixed effects across all clusters, while the sandwich estimator gives appropriate inferences for either fixed effects or random effects. However, inference using the standard Wald test with the sandwich estimator requires that each parameter is estimated using information from a large number of clusters. Since our example violates this condition, we introduce two modifications to the standard Wald test. First, we reduce the bias of the empirical variance estimator (the middle of the sandwich) by using standardized residuals. Second, we approximately account for the variance of these estimators by using the t-distribution instead of the normal distribution, where the degrees of freedom are estimated using Satterthwaite's approximation. Through simulations, we show that these sandwich estimators perform almost as well as classical estimators when the true effects are fixed and much better than the classical estimators when the true effects are random. We achieve simulated nominal coverage for these sandwich estimators even when some parameters are estimated from a small number of clusters.     

Individual-group discontinuity as a function of fear and greed.

Two studies tested the schema-based distrust interpretation of the tendency of intergroup relations to be more noncooperative (or competitive) than interindividual relations. According to this interpretation, anticipated competitiveness rationally leads to noncooperativeness or defensive withdrawal. Thus, the postulated motivation is fear of the group's competitive intent. Study 1 was a nonexperimental investigation in which discussion of distrust of another group was assessed and correlated with the number of cooperative choices. As predicted, the greater the within-group discussion of distrust for the other group, the less the number of cooperative choices. Study 2 was an experimental investigation that included as independent variables intergroup versus interindividual relations and PDG matrix versus PDG-Alt matrix (PDG matrix plus a third Alt or withdrawal choice producing intermediate outcomes regardless of the opponent's choice). As predicted, there were more withdrawal choices on the PDG-Alt matrix for groups than for individuals. However, it was still found that on the PDG-Alt matrix (where a safe withdrawal choice is possible), groups competed more than individuals. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Disturbances of affective prosody in patients with schizophrenia; a cross sectional study

The solutions of nine alpha, omega-bis(3-alkyl(aryl)-4,5-dihydro-1H-1,2,4-triazol-5-on-4-yl) alkanes were titrated with tetrabutylammoniumhydroxide (TBAH) in methanol, using potentiometric methods. The half neutralization potentials values were found for all cases. Potentiometric titration curves of compounds in methanol with 0.03 M TBAH are similar to those of weak acids obtained in aqueous media with strong bases. Methanol is found to be a suitable medium for the weakly acidic compounds titrated since they are poorly dissolved in other organic solvents. A comparison among the compounds having the same alkyl chains between the two ring systems has shown that basicity increases and acidity decreases as the size of alkyl chains increases. However, the compound with a substituted phenyl group was found to be the most acidic one among the examined compounds indicating that phenyl group donates ring electrons less effectively to the system. This can be attributed to the stability of the benzene ring.     

Inference using conditional logistic regression with missing covariates

When there are many nuisance parameters in a logistic regression model, a popular method for eliminating these nuisance parameters is conditional logistic regression. Unfortunately, another common problem in a logistic regression analysis is missing covariate data. With many nuisance parameters to eliminate and missing covariates, many investigators exclude any subject with missing covariates and then use conditional logistic regression, often called a complete-case analysis. In this article, we derive a modified conditional logistic regression that is appropriate with covariates that are missing at random. Performing a conditional logistic regression with only the complete cases is convenient with existing statistical packages, but it may give bias if missingness is not completely at random.     

Stimulus-driven attentional capture by a static discontinuity between perceptual groups.

After C. L. Folk, R. W. Remington, and J. C. Johnston (1992) proposed their contingent-orienting hypothesis, there has been an ongoing debate over whether purely stimulus-driven attentional capture can occur for visual events that are salient by virtue of a distinctive static property (as opposed to a dynamic property such as abrupt onset). The present study identified 3 methodological criteria for establishing that attentional capture is stimulus driven and not contingent on top-down attentional control settings. In 5 experiments, attentional capture occurred for a static discontinuity at the boundary between one group of homogeneous items (red Xs) abutted next to a group of homogeneous items that were featurally different (green Xs) within a single row. Experiment 1 intentionally violated one of the criteria for demonstrating stimulus-driven capture so as to establish that contingent attentional capture can occur for this novel type of static cue. In the remaining 4 experiments, even with all 3 criteria for stimulus-driven capture partially or completely satisfied, the static discontinuity captured attention. These attentional capture effects are the first to be obtained when all 3 criteria for establishing that they are purely stimulus driven have been satisfied. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Within-category discontinuity interacts with verbal rule complexity in perceptual category learning.

A test of the predicted interaction between within-category discontinuity and verbal rule complexity on information-integration and rule-based category learning was conducted. Within-category discontinuity adversely affected information-integration category learning but not rule-based category learning. Model-based analyses suggested that some information-integration participants improved performance by recruiting more "units" in the discontinuous condition. Verbal rule complexity adversely affected rule-based category learning but not information-integration category learning. Model-based analyses suggested that the rule based effect was on both decision criterion learning and variability in decision criterion placement. These results suggest that within-category discontinuity and decision rule complexity differentially impact information-integration and rule-based category learning and provide information regarding the detailed processing characteristics of these two proposed category learning systems. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Toward a cDNA map of the human genome

Advances in the Human Genome Project are shaping the strategies for identifying the 50,000-100,000 human genes. High-resolution genetic maps of the human genome combined with sequencing herald an era of rapid regional definition of disease genes. However, only once their chromosome band location is known will the systematic partial sequencing of thousands of random cDNA clones provide the reagents for teh rapid assessment of the genes responsible for the inherited disorders. We now present an approach to the rapid determination of map position and therefore to the creation of a transcribed map of the human genome. Sensitive fluorescence in situ hybridization has been combined with high-resolution chromosome banding and random cDNA sequencing to map 41 cDNAs with an average insert size of <2 kb to single human chromosome bands. The result provide 15 new genes, with database and functional information, as candidates for human disease. These include the large extracellular signal-related kinase (HUMERK), the ERK activator kinase (PRKMK1), a new member of the RAS oncogene family, protein phosphatase 2 regulatory subunit B alpha isoform (PPP2R2A), and a novel human gene with very high homology to a plant membrane transport family. Further, an analysis of expressed genes associated with pseudogenes showed that by using these techniques, it is possible to detect accurately the transcribed locus within a multigene or processed pseudogene family in most cases. These findings suggest that direct cDNA mapping using fluorescence in situ hybridization provides an accurate and rapid approach to the definition of a transcribed map of the human genome. This low-cost, high-resolution (2-5 Mb) mapping greatly enhances the speed with which these genes can be subsequently assigned to contigs. This assignment provides a necessary first step in understanding the relationship of the genes to both acquired and inherited human diseases.