Consequences of effect size heterogeneity for meta-analysis: a Monte Carlo study

Methods of estimating the pooled effect size under meta-analysis: A comparative appraisal

ABSTRACTThe power system state estimator based on the support vector machine (SVM) and the weighted least squares (WLS) method is presented in the paper. The WLS provides state estimations necessary for creating SVM model which is then used for state estimation. The developed algorithm was tested on the IEEE systems, and the performance indicators were calculated in order to compare the accuracy of estimation and the measurement error filtering. The results indicate that the proposed hybrid model outperforms the classical WLS-based state estimation in terms of accuracy and improves measurement error filtering in comparison to the classical estimator.

Meta-analysis is a statistical method for synthesizing quantitative data from multiple related studies, yet heterogeneity among studies often complicates interpretation. Meta-regression extends this approach by incorporating study-level covariates to explain variations in outcomes. With the global increase in depression, Acceptance and Commitment Therapy(ACT) has attracted attention as an effective psychological intervention. Therefore, a deeper understanding of the factors that influence its effectiveness across studies is needed. However, to date, only a few meta-analyses have quantitatively examined moderator variables that influence ACT outcomes using a random effects meta-regression approach. This study aims to fill this gap. This study estimated the model parameters using the Weighted Least Squares (WLS) method. Thirty-three published studies testing the effectiveness of ACT in reducing depression were collected from PubMed, Google Scholar, and Science Direct. The homogeneity test results showed significant heterogeneity, supporting the use of a random effects model. The combined effect size of -0.321 indicates that ACT significantly reduces depression levels compared to the control group. Meta-regression analysis revealed that variation in effect size was significantly influenced by differences in the average age of patients and duration of therapy. These findings provide new insights into the conditions and characteristics that make ACT therapy more effective.

This systematic review and meta-analysis aimed to compare the efficacy of intralesional platelet-rich plasma and intralesional steroids in managing oral lichen planus, focusing on pain and lesion size reduction. A systematic search of MEDLINE (via PubMed), Web of Science, Scopus, and Embase was conducted up to May 30, 2024, to identify randomized controlled trials (RCTs) comparing the efficacy of intralesional platelet-rich plasma (PRP) and intralesional steroids in managing oral lichen planus (OLP). The study adhered to a predefined PICOS framework: (P) adults diagnosed with OLP, (I) PRP application, (C) steroid application, (O) outcomes related to pain reduction and lesion size, and (S) RCTs. The analysis utilized both random-effects and Weighted Least Squares (WLS) models to account for heterogeneity and assess publication bias. Seven articles met the eligibility criteria, of which six were included in the meta-analysis, encompassing 141 participants. While the analysis was limited by a small sample size, significant heterogeneity, and a lack of long-term follow-up, analysis with random-effects and WLS model provided a robust comparative evaluation of PRP and steroids for OLP treatment. Pain Reduction: The pooled effect size was -0.34 (95% CI: -0.94 to 0.27; p = 0.274), indicating a moderate, non-significant preference for steroids over PRP. Moderate to high heterogeneity (I² = 68.516%) suggested substantial variability in effect sizes. Lesion Size Reduction: The pooled effect size was -0.10 (95% CI: -0.58 to 0.39; p = 0.697), favoring PRP slightly but without statistical significance. Low to moderate heterogeneity (I² = 31.42%) was observed. PRP demonstrated comparable efficacy to steroids in reducing pain and lesion size in OLP patients, with the added advantage of fewer side effects. However, significant heterogeneity and publication bias necessitate further high-quality research to substantiate these findings.

In experimental design applications unbiased estimators si 2 of the variances σi 2 are possible. These estimators may be used in Weighted Least Squares (WLS) when estimating the parameters β. The resulting small-sample behavior is investigated in a Monte Carlo experiment. This experiment shows that an asymptotically valid covariance formula can be used if si 2 is based on, say, at least 5 observations. The WLS estimator based on estimators si 2 gives more accurate estimators of β, provided the σi 2 differ by a factor, say, 10. Keywords: generalized least squaresexperimental errorMonte Carlo

In the least-squares fitting of data of varying precision to functions that are algebraically linear in the adjustable parameters, the correct weights for obtaining optimal results are the pointwise inverse variances σi-2. These have often been obtained from the sampling statistics of replicate measurements of the uncertain quantity y, a procedure that has come to define weighted least-squares (WLS) calibration fitting. Unfortunately, such variance estimates are notoriously imprecise, and in a long-ago Monte Carlo study involving a linear response function and weights ranging over a factor of 5, ordinary least squares (OLS) outperformed WLS for almost every tested set of calibration x values and replicates. But there is a better way of obtaining the weights: variance-function estimation. This method relies on the fact that the variances for most physical data follow simple, smooth functions having two or three terms proportional to a constant, y, and y2. Since the replicate data are used collectively to estimate the variance function, the number of statistical degrees of freedom is large enough to achieve good precision in the calibration weights. When variance-function weighting is included in the aforementioned OLS-vs-WLS comparisons, it wins over OLS for almost every combination with m = 4-8 x values and n = 3-5 replicates. Further, of the four ways of weighting in the variance-function fitting, the iterative reweighting method gives the best results, in support of earlier findings.

ABSTRACTSeveral approaches exist for estimating the derivatives of observed data for model exploration purposes, including functional data analysis (FDA; Ramsay & Silverman, 2005), generalized local linear approximation (GLLA; Boker, Deboeck, Edler, & Peel, 2010), and generalized orthogonal local derivative approximation (GOLD; Deboeck, 2010). These derivative estimation procedures can be used in a two-stage process to fit mixed effects ordinary differential equation (ODE) models. While the performance and utility of these routines for estimating linear ODEs have been established, they have not yet been evaluated in the context of nonlinear ODEs with mixed effects. We compared properties of the GLLA and GOLD to an FDA-based two-stage approach denoted herein as functional ordinary differential equation with mixed effects (FODEmixed) in a Monte Carlo (MC) study using a nonlinear coupled oscillators model with mixed effects. Simulation results showed that overall, the FODEmixed outperformed both the GLLA and GOLD across all the embedding dimensions considered, but a novel use of a fourth-order GLLA approach combined with very high embedding dimensions yielded estimation results that almost paralleled those from the FODEmixed. We discuss the strengths and limitations of each approach and demonstrate how output from each stage of FODEmixed may be used to inform empirical modeling of young children’s self-regulation.

Local Extinction and Recolonization, Species Effective Population Size, and Modern Human Origins

The purpose of this study was to compare the most demanding scenarios (MDS) encountered by professional basketball players across game quarters and playing positions during official match-play. Ten professional bas- ketball players were monitored during 11 matches using a local positioning system. Peak physical demands were measured via total distance, distance >18 and >21 km·h -1 , number of sprints >18 and >21 km·h -1 , and number of accelerations and decelerations >2 and >3 m∙s -2 captured over 30, 60, 120, 180, and 300-s rolling averages. Linear mixed models and effect sizes (ES) were used to compare MDS encountered between game quarters and playing positions. Between Quarters 1 and 2, there was a reduction in the total distance (ES = 0.64-1.39) for all playing groups along with a reduction in distance >21 km·h -1 in centres (ES = 0.77-0.81) and a reduction in accelerations and decelerations >2 m∙s -2 in guards (ES = 0.66-0.78) across longer sample periods (180-300-s). Between Quarters 1 and 4, reductions in the total distance were evident for forwards and centres (ES = 0.64-0.91) as well as distance and sprints >21 km·h -1 in centres (ES = 0.64-0.97). Regarding positional differences, guards and forwards covered a higher total distance than centres across most quarters and sample periods (ES = 0.22-1.44). Our data suggest a decrease in MDS with game progression in basketball. In addition, MDS appear to be duration-specific and posi- tion-dependent in basketball. Therefore, practitioners should consider these differences in MDS based on game quarters and positional demands to optimise individual and team performance.

PurposeThe purpose of this paper is to investigate convergence or divergence trends at global scale.Design/methodology/approachThe paper questions the methodology and findings of the conventional convergence literature using linear OLS models. It introduces polynomial (quadratic) weighted least square (WLS) regression analysis to explore whether a number of economic performance indicators follow a non‐linear pattern of change.FindingsThe results indicate the formation of two groups in the world: a convergence one, including countries with low to medium‐high development levels, and a divergence one including countries with medium‐high to very high development levels.Research limitations/implicationsData availability after 1990 (for the composite indicators).Practical implicationsThe findings shed light on important issues, such as the decrease of economic disparities between countries, the prospects for global economic convergence, and the development of a more equal world. Apart from obvious policy implication such findings are also of theoretical significance, providing a basis to check (indirectly) the validity of alternative growth theories.Originality/valueThis is the first paper (to the authors' knowledge) that explores world convergence/divergence employing quadratic WLS regression analysis with a number of economic indicators. WLS regressions enable the removal of the impact of country size on results, whereas non‐linear modelling allows the possibility of multiple equilibria and different development trajectories to be taken into account. Finally, the employment of various economic‐performance indicators (simple and composite) works as a cross‐check of validity for the results provided.

Purpose: To investigate the strengths and limitations of different indicators to measure physical load. Furthermore, indicators were evaluated for discrimination between performance levels and playing positions. Methods: Ninety positional match files from 70 elite players and 91 match files from 69 subelite players were collected during 14 official under-18 matches using a local position measurement system. Indicators are calculated from speed, absolute acceleration (acc-abs), or percentage acceleration (acc-%). The acc-% describes the level of acceleration depending on the maximal voluntary acceleration (amax) for each initial running speed. Effect sizes (ES) were used to determine discriminative ability. Results: The number of high accelerations largely depended on the method (absolute threshold [>3m·s-2 and >4m·s-2] 120 and 59 efforts; high percentage threshold [>75% amax] 84 efforts). Only a small number of highly accelerated efforts reached speeds considered high-speed running (>19.8km·h-1: 32.6%). More high acc-% exists from initial running speed >2m·s-1 (23.0) compared with acc-abs (>3m·s-2 14.4, >4m·s-2 5.9). Elite players achieve higher values in most performance indicators, with ES being highest for the number of high acc-% (ES = 0.91) and high acc-abs (>3m·s-2 ES = 0.86, >4m·s-2 ES = 0.87), as well as for covered distance in jogging (ES = 0.94). Conclusions: Estimated physical load, discriminative ability of physical indicators, and positional requirements largely depend on the applied method. A combination of speed-based and acc-% methods is recommended to get a comprehensive view.

INSTRUMENTAL MEASUREMENTS: Resistance to Tear: An Instrumental Measure of the Cohesive Properties of Muscle Food Products. (Res. Note) J. H. MacNeil and M. G. MastINSTRUMENTAL MEASUREMENTS: A Small Sample Back Extrusion Test for Measuring Texture of Cooked Rice. V. G. Reyes, Jr. and V. K. JindalINSTRUMENTAL MEASUREMENTS: The Wheat Research Institute Chomper, an Instrument that Measures Crumb Flexibility. D. W. Baruch and T. D. AtkinsINSTRUMENTAL MEASUREMENTS: Using the Wheat Research Institute Chomper to Assess Crumb Flexibility of Staling Bread. D. W. Baruch and T. D. AtkinsINSTRUMENTAL MEASUREMENTS: Rapid Testing Method for Characterizing the Rheological Behavior of Gelatinizing Corn Starch Slurries. (Note) J. F. Steffe, M. E. Castell Peres, K. J. Rose and M. E. ZabickINSTRUMENTAL MEASUREMENTS: Rheological Changes of Fortified Wheat and Corn Flour Doughs with Mixing Time. L. L. NavickisINSTRUMENTAL MEASUREMENTS: Comparison of Starch Pasting Properties in the Brabender Viscoamylograph and the Rapid Visco‐Analyzer. L. B. Deffenbaugh and C. E. WalkerINSTRUMENTAL MEASUREMENTS: Measuring Kernel Hardness Using the Tangential Abrasive Dehulling Device. J. W. Lawton and J. M. FaubionINSTRUMENTAL MEASUREMENTS: Bingham Award Lecture—1989. The Role of Instrument Inertia in Controlled‐Stress Rheometers. 1. M. KriegerINSTRUMENTAL MEASUREMENTS: Technical Note: Strain‐Measurement Error in a Constant‐Stress Rheometer. D. W. Giles and M. M. DennINSTRUMENTAL MEASUREMENTS: Technical Note: Strain‐Measurement Error in a Constant‐Stress Rheometer. D. W. Giles and M. M. DennINSTRUMENTAL MEASUREMENTS: Characteristics of a Modified Wisconsin Breakage Tester. (Technical Note) H. J. Lyon, S. G. Schmitt, C. J. Bern and C. R. Hurburgh, Jr.INSTRUMENTAL MEASUREMENTS: Impact Parameters Related to Post Harvest Brusing of Apples. G. H. Brusewitz and J. A. BartschINSTRUMENTAL MEASUREMENTS: Characterization of the Perceived Texture of Thickened Systems by Dynamic Viscosity Measurements. R. K. Richardson, E. R. Morris, S. B. Ross‐Murphy, L. J. Taylor and I.C.M. DeaGENERAL PRINCIPLES: Modelling of Flow Properties of Starch Pastes Prepared by Different Procedures. M. HarrodGENERAL PRINCIPLES: Apparent Concentration: A Method to Predict the Flow Properties of Viscous Foods for Process Applications. M. HarrodGENERAL PRINCIPLES: Time‐Dependent Flow Behavior of Starch Pastes, With Food Process Applications. M. HärrödGENERAL PRINCIPLES: Intrinsic Viscosities and Apparent Specific Volumes of Amino Acids and Sugars. ‘Effective Size’ and Taste of Sapid Molecules. S. E. Kemp and G. G. Birch, M. 0. Portmann and M. MathlouthiGENERAL PRINCIPLES: Effect of Succinylation on the Functional and Physiochemical Properties of Bovine Serum Albumin. Margaret C. Murphy and Nazlin K. HowellGENERAL PRINCIPLES: Effect of Extrusion Temperature on Physico‐Chemical Properties and Biological Value of Soybean‐Protein. E. Horváth, J. Petres, É, Gelencsér and B. CzukorSENSORY EVALUATION: Adaptation of Collaborative Methodology to the Sensory Evaluation of Foods by Scoring. P. MolnárFACTORS AFFECTING TEXTURE: Factors Related to Longissimus Tenderness Among Alternative Methods of Lean Beef Production. W. Vanderwert, P. J. Bechtel, D. L. DeVol, F. K. McKeith, L. L. Berger and R. D. ShanksFACTORS AFFECTING TEXTURE: Predicting the Effect of Drying Conditions on the Textural Properties of Roasted Peanuts. Y.‐C. HungFACTORS AFFECTING TEXTURE: Effect of Maturity and Storage Time on the Bruise Susceptibility of Peaches (cv. Red GIobe).Y.‐C. Hung and S. E. PrussiaFACTORS AFFECTING TEXTURE: Mechanical Properties of Kappa Carrageenan‐Locust Bean Gum Mixed Gels with Added Sucrose. S. M. Fiszman and L. DuranFACTORS AFFECTING TEXTURE: Effect of Hydrolysis of Lactose and Sucrose on Firmness of Ice Cream. J. B. Lindamood, D. J. Grooms and P.M.T. HansenFACTORS AFFECTING TEXTURE: Effect of Chemical Additives on Toughening of Fillets of Frozen Alaska Pollack (Therugru chulcogrumrnu). D. J. Krueger and 0. R. FennemaFACTORS AFFECTING TEXTURE: Effect of Mixing and Moisture Modification on Toughening and Dimethylamine Formation in Alaska Pollock Mince during Frozen Storage at −10°C. G. Boer and O. FennemaFACTORS AFFECTING TEXTURE: Rheological Properties of Heat‐Induced Ovalbumin Gels Prepared by Two‐step and One‐step Heating Methods. N. Kitabatake, Y. Tani and E. DoiFACTORS AFFECTING TEXTURE: Physical and Sensory Evaluation of Lean White Cakes Containing Substituted Fluffy Cellulose. E. B. Fondroy, P. J. White and K. J. Prusa

Researchers have increasingly investigated sources of variance in reading comprehension test scores, particularly with students with reading difficulties (RD). The purpose of this meta-analysis was to determine if the achievement gap between students with RD and typically developing (TD) students varies as a function of different reading comprehension response formats (e.g., multiple choice, cloze). A systematic literature review identified 82 eligible studies. All studies administered reading comprehension assessments to students with RD and TD students in Grades K–12. Hedge’s g standardized mean difference effect sizes were calculated, and random effects robust variance estimation techniques were used to aggregate average weighted effect sizes for each response format. Results indicated that the achievement gap between students with RD and TD students was larger for some response formats (e.g., picture selection ES g = −1.80) than others (e.g., retell ES g = −0.60). Moreover, for multiple-choice, cloze, and open-ended question response formats, single-predictor metaregression models explored potential moderators of heterogeneity in effect sizes. No clear patterns, however, emerged in regard to moderators of heterogeneity in effect sizes across response formats. Findings suggest that the use of different response formats may lead to variability in the achievement gap between students with RD and TD students.

Quaternary ammonium compounds (QACs) have been used as disinfestants in plant production systems since the late 20th century. In studies on the control of fungal pathogens in agricultural and horticultural crop production systems, the efficacy of QAC disinfestants is variable, ranging from very high to ineffective. A systematic review and meta-analysis were performed to establish and understand how pathogen- and application-related factors influenced product efficacy. The meta-analysis was based on 124 studies involving 14 fungal plant pathogen genera, eight target materials, and four generations of QAC products that contained different mixtures of active ingredients. A significant (P < 0.0001) reduction in either disease intensity or propagule viability resulted following disinfestation using QAC products. Hedges' g standardized mean difference (g̅+) across the studies was 2.16, indicating that QACs, on average, were highly effective against fungal pathogens. Heterogeneity was significant (P < 0.0001), indicating that effect sizes (g) were not representative of a common mean effect size and supported selection of a random effects model. In all, 78.5% of the observed variance consisted of variance in true effects with a high estimate of between-study variability (τ2 = 2.15). For fungal genus, subgroup g̅+ for genera Pseudonectria and Calonectria was significantly (P < 0.0038) higher than for all other genus subgroups, except Fusarium. For target materials, subgroup g̅+ for solution, cloth, plant, and metal were significantly (P > 0.0071) higher than for inorganic material or wood. For product generation, subgroup g̅+ for fifth-generation products was significantly (P > 0.0071) higher than for fourth-, third-, and second-generation products. Dose and time accounted for only 8 and 4%, respectively, of the true variance in effect sizes in the regression model dose, time, and dose-time (P = 0.0004). Genus accounted for 40 and 51% of the true variance in effect sizes in the regression models dose and genus (P = 0.0008) and time and genus (P = 0.0007), respectively. Target material accounted for 18 and 19% of the true variance in effect sizes in the regression models dose and target (P = 0.0001) and time and target (P = 0.0001), respectively. QAC product generation accounted for 24 and 21% of the true variance in effect sizes in the regression models dose and QAC generation (P = 0.0034) and time and QAC generation (P = 0.0189), respectively. These results show that the current recommended rates for dose and contact time are generally expected to result in effective disinfestation for commercial QAC products. However, the efficacy against fungal plant pathogens is likely to be influenced by the fungal genus and target being treated and the generation of the QAC product that is used for disinfestation.

Visual Analogue Scale has higher assay sensitivity than WOMAC pain in detecting between-group differences in treatment effects: a meta-epidemiological study