An application of hierarchical linear models to meta-analysis in nursing research

OBJECTIVES: To compare the effect of using regions of interest (ROIs) of different size and shape on the fractal index of alveolar bone. STUDY DESIGN: Two sets of clinical posterior bitewing radiographs were used to calculate the fractal index (S). Two comparisons were made. First, S was calculated from large interdental ROIs that included small amounts of root structure and compared with S from small ROIs that included no root structures. Then S was calculated from large interdental ROIs (similar to those used for the first set) and compared with S calculated from ROIs that included nearly all of the mandibular alveolar bone (and adjacent root) present on the bitewing. RESULTS: For the first comparison, paired t-tests showed that fractal indices calculated with the large ROIs were significantly different from the respective indices calculated from the small ROIs (P < 0.001). For the second comparison, the fractal indices calculated from the large quadrant ROIs were not significantly different from those calculated from the large ROIs (P = 0.120). CONCLUSION: ROI size and shape may affect the results of fractal analysis of alveolar bone.     

Estimating coefficients in linear models: It don't make no nevermind.

Reviews the literature and shows that under very general circumstances coefficients in multiple regression models can be replaced with equal weights with almost no loss in accuracy on the original data sample. It is then shown that these equal weights will have greater robustness than least squares regression coefficients. The implications for problems of prediction are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Program evaluation using hierarchical linear modeling with curriculum-based measurement reading probes.

Used hierarchical linear modeling (HLM) to evaluate 99 ethnically diverse 2nd-Graders' growth in oral reading fluency using curriculum-based measurement (CBM) over the course of the school year. Five statistical advantages of the use of HLM are described and four are highlighted in this example: (a) The potential for improved accuracy in slope estimation of the empirical Bayes method used in HLM; (b) HLM provided the flexibility to test the effects of independent variables on the initial status of the slope and the change in slope over time; (c) HLM provided an analysis of the nested effects of the classroom on individual performance; and (d) HLM provided a statistical test for both group effects and individual variation in reading growth. Results showed 1st-Grade reading performance significantly predicted initial 2nd-Grade reading performance. Individual growth curves of 28 students who attended summer school showed that all of them made statistically significant growth in reading ability. 71% of the summer school students scored between the 1st and 2nd quartile of the spring, norming distribution at the end of summer school. This example shows that the use of HLM with normative CBM reading probes provided additional information not available with other statistical techniques. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Comment on "Estimating coefficients in linear models: It don't make no nevermind."

Suggests that H. Wainer's (see record 1976-29043-001) equal weights theorem is in error and that the loss in explained variance when replacing optimal least squares weights by equal coefficients is twice as great as contended. An alternative formulation and proof of the theorem is developed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Interpretive work in short-term individual psychotherapy: An analysis using hierarchical linear modeling.

"Work" and "resistance" responses to interpretation in short-term individual (STI) psychotherapy were examined using a hierarchical linear modeling (HLM) procedure. The relationships between interpretation characteristics and patient responses within therapy were considered. Process data were drawn from 60 STI therapy cases, 30 patients with low quality of object relations (QOR), and 30 patients with high QOR. In 4 instances, the relationships between technique and response were found to vary significantly across cases. One was identified for low QOR patients, and 3 were identified for high QOR patients. Individual differences in initial disturbance and outcome were used to account for the variation of technique-response relationships. Significant findings were limited to the high QOR sample. Initial disturbance was directly related to work in response to a transference oriented approach. The transference focus-work relationship was found to be inversely related to outcome. The results extend previous findings regarding transference technique in STI therapy with high QOR patients. Through capitalizing on within-case variation, HLM can be used to illuminate process-outcome relationships in psychotherapy. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Regression models for recurrent event data: parametric random effects models with measurement error

Statistical methodology is presented for the statistical analysis of non-linear measurement error models. Our approach is to provide adjustments for the usual maximum likelihood estimators, their standard errors and associated significance tests in order to account for the presence of measurement error in some of the covariates. We illustrate the technique with a mixed effects Poisson regression model for recurrent event data applied to a randomized clinical trial for the prevention of skin tumours.     

A note on the phenomenological coefficients for atom transport in a dilute random alloy

Manning's expressions for the phenomenological atomic transport coefficients for a random alloy are known from computer simulations to be remarkably accurate for concentrated alloys. These expressions for a binary alloy have been evaluated in the dilute solution limit (i.e. correct to first order in the solute concentration) and compared with exact and accurate numerical results for this limit obtained by specializing known results for vacancy models to the random alloy case. It is found that three of the four independent coefficients are given exactly by Manning's expressions. Although there are no exact results for the fourth coefficient (i.e. for the “enhancement factor” for diffusion of tracers of the solvent species), the trend of this quantity as a function of the ratio of the jump frequencies of the components is in agreement with earlier numerical calculations for both f.c.c. and b.c.c. lattice models. Anomalously in our view, the trend predicted by some more recent numerical calculations [S. Ishioka and M. Koiwa, Phil. Mag. A50, 505 (1984)] is in the opposite sense.     

Examining psychokinesis: The interaction of human intention with random number generators--A meta-analysis.

Séance-room and other large-scale psychokinetic phenomena have fascinated humankind for decades. Experimental research has reduced these phenomena to attempts to influence (a) the fall of dice and, later, (b) the output of random number generators (RNGs). The meta-analysis combined 380 studies that assessed whether RNG output correlated with human intention and found a significant but very small overall effect size. The study effect sizes were strongly and inversely related to sample size and were extremely heterogeneous. A Monte Carlo simulation revealed that the small effect size, the relation between sample size and effect size, and the extreme effect size heterogeneity found could in principle be a result of publication bias. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Estimating unconscious processes: Implications of a general class of models.

Jacoby and his colleagues (e.g., L.L. Jacoby, J.P. Toth, & A.P. Yonelinas, see record 1993-36166-001) have shown how the mnemonic contributions of conscious recollection and unconscious familiarity can be separated using a process-dissociation procedure based on a comparison of tasks in which consciously recollected material is to be included in vs. excluded from the responses. However, the estimate of unconscious familiarity depends on the assumptions of the model. This article describes a more general class of models in which the ratio of overlap between conscious and unconscious processes remains fixed and shows that this class of models (which includes the redundancy and exclusivity models as extreme cases) yields improbable results. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Intralist distractors and recall direction: Constraints on models of memory for serial order.

The serial-order version of the theory of distributed associative memory (TODAM; S. Lewandowsky and B. B. Murdock [see PA, Vol 76:14457]) predicts that disuption of forward serial recall should leave backward recall largely unaffected. This article reports 4 experiments in which the effects of an intralist distractor task were compared for forward and backward serial recall. Regardless of whether Ss could anticipate recall direction at study, the distractor task was found to disrupt forward but not backward recall. Although the existence of that dissociation had been predicted by TODAM, the theory was unable to provide a quantitative account of the data. Instead the authors provide a retrieval-based account within the framework of temporal distinctiveness theory. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Therapy of shock with naloxone: a meta-analysis

OBJECTIVES: To evaluate the effectiveness of naloxone in human shock; and to estimate the methodologic quality of the clinical trials. DATA SOURCES: Computerized bibliographic search on MEDLINE covering the period from January 1979 to July 1996, review of references of all papers found on the subject, and contact with primary investigators of eligible studies. STUDY SELECTION: To be included in this study, a paper should be a randomized, clinical trial published in a peer-reviewed journal evaluating naloxone in human shock, regardless of the patient's age (adult, child, neonate). Three independent readers reviewed 61 human publications and selected five clinical trials. Overall agreement on study selection was perfect (concordance: 100%). We excluded a posteriori two studies whose authors were unable to provide us with the raw data to complete contingency tables. This meta-analysis deals with three studies including 61 patients with septic shock. DATA EXTRACTION: Three independent reviewers extracted data on study design, intervention, outcome, and methodologic quality. The intraclass correlation coefficient was 0.7. The quality score of each study was 48, 60, and 61, on a scale of 104. DATA SYNTHESIS: Naloxone therapy was associated with statistically significant hemodynamic improvement (typical odds ratio: 0.241; 95% confidence interval: 0.08 to 0.68). The overall effect size was 0.89. However, a publication bias was possible. The case fatality rate was not decreased by naloxone (typical odds ratio: 0.60; 95% confidence interval: 0.21 to 1.67); a chi-square analysis detected significant heterogeneity for the latter outcome (p < .05). CONCLUSIONS: Naloxone improves blood pressure. However, the clinical usefulness of naloxone to treat shock remains to be determined and additional randomized clinical trials are needed to assess its usefulness.