Number Of Moderator Variables Research Articles

A Meta‐Analysis of Hybrid Language Instruction and Call for Future Research

AbstractIn this meta‐analysis, we estimate the effectiveness of hybrid language instruction overall and across a number of moderator variables by aggregating effect sizes from 11 studies with 34 samples. Results suggest hybrid language instruction can be just as effective as traditional face‐to‐face (f2f) instruction, as indicated by the negligible differences between hybrid courses and traditional f2f courses (d = .14). Furthermore, studies employing within‐group designs indicate that students in hybrid language classes can improve their language skills considerably (d = 1.47). This is a positive finding given that many institutions have experienced a surge in hybrid teaching due to the COVID‐19 pandemic. We also report on a number of moderator variables that can impact the effectiveness of hybrid language courses, including (a) the amount of reduction in f2f time, (b) the use of online activities provided by textbook publishers, (c) the use of a learning management system, (d) advances in digital technologies, (e) the targeted language skills (e.g., speaking, writing), and (f) whether the data come from initial or subsequent iterations of a hybrid course. Additionally, we offer directions for future research regarding the substantive and methodological issues in the hybrid language instruction domain.

A subdivision algorithm to reason on high-degree polynomial constraints over finite domains

This paper proposes an algorithm to reason on constraints expressed in terms of polynomials with integer coefficients whose variables take values from finite subsets of the integers. The proposed algorithm assumes that an initial approximation of the domains of variables is available in terms of a bounding box, and it recursively subdivides the box into disjoint boxes until a termination condition is met. The algorithm includes three termination conditions that allow using it for three related reasoning tasks: constraint satisfaction, enumeration of solutions, and hyper-arc consistency enforcement. Considered termination conditions are based on suitable lower and upper bounds for polynomial functions over boxes that are determined using new results proved in the paper. The algorithm is particularly appropriate to reason on high-degree polynomial constraints because the proposed method to determine lower and upper bounds can outperform alternative methods when high-degree polynomials in a moderate number of variables are considered.

"Validity of content-based techniques to distinguish true and fabricated statements: A meta-analysis": Correction to Oberlader et al. (2016).

Reports an error in "Validity of content-based techniques to distinguish true and fabricated statements: A meta-analysis" by Verena A. Oberlader, Christoph Naefgen, Judith Koppehele-Gossel, Laura Quinten, Rainer Banse and Alexander F. Schmidt (Law and Human Behavior, 2016[Aug], Vol 40[4], 440-457). During an update of this meta-analysis it became apparent that one study was erroneously entered twice. The reduced data of k = 55 studies was reanalyzed after excluding the unpublished study by Scheinberger (1993). The corrected overall effect size changed at the second decimal: d = 1.01 (95% CI [0.77, 1.25], Q = 409.73, p < .001, I² = 92.21) and g = 0.98 (95% CI [0.75, 1.22], Q = 395.49, p < .001, I² = 91.71%), k = 55, N = 3,399. This small numerical deviation is negligible and does not change the interpretation of the results. Similarly, results for categorial moderators changed only numerically but not in terms of their statistical significance or direction (see revised Table 4). In the original meta-analysis based on k = 56 studies, unpublished studies had a larger effect size than published studies. Based on k = 55 studies, this difference vanished. Results for continuous moderators also changed only numerically: Q-tests with mixed-effects models still revealed that year of publication (Q = 0.06, p = .807, k = 55) as well as gender ratio in the sample (Q = 1.28, p =.259, k = 43) had no statistically significant influence on effect size. In sum, based on the numerically corrected values our implications for practical advices and boundary conditions for the use of content-based techniques in credibility assessment are still valid. The online version of this article has been corrected. (The following abstract of the original article appeared in record 2016-21973-001.) Within the scope of judicial decisions, approaches to distinguish between true and fabricated statements have been of particular importance since ancient times. Although methods focusing on "prototypical" deceptive behavior (e.g., psychophysiological phenomena, nonverbal cues) have largely been rejected with regard to validity, content-based techniques constitute a promising approach and are well established within the applied forensic context. The basic idea of this approach is that experience-based and nonexperience-based statements differ in their content-related quality. In order to test the validity of the most prominent content-based techniques, criteria-based content analysis (CBCA) and reality monitoring (RM), we conducted a comprehensive meta-analysis on English- and German-language studies. Based on a variety of decision criteria, 55 studies were included revealing an overall effect size of g = 0.98 (95% confidence interval [0.75, 1.22], Q = 395.49, p < .001, I² = 91.71%, N = 3,399). There was no significant difference in the effectiveness of CBCA and RM. Additionally, we investigated a number of moderator variables, such as characteristics of participants, statements, and judgment procedures, as well as general study characteristics. Results showed that the application of all CBCA criteria outperformed any incomplete CBCA criteria set. Furthermore, statement classification based on discriminant functions revealed higher discrimination rates than decisions based on sum scores. All results are discussed in terms of their significance for future research (e.g., developing standardized decision rules) and practical application (e.g., user training, applying complete criteria set). (PsycINFO Database Record (c) 2019 APA, all rights reserved).

The effectiveness of L2 vocabulary instruction: a meta-analysis

The purpose of the present meta-analysis is to the investigate the overall effectiveness of L2 vocabulary instruction and to find the moderator variables affecting its effectiveness. By defining a rigorous inclusion and exclusion criteria, a total number of 16 primary studies (N = 1008), 7 published and 9 Ph.D. dissertations, were included. Under Random-Effects Model, the overall effect size of (d = 0.80) was observed. After conducting Q test of heterogeneity, a number of moderator variables were examined; context of instruction, publication type, age and L2 learners’ proficiency level. It was found that (a) studies conducted in foreign language contexts generated larger effect sizes than ones conducted in SL contexts.(b)intermediated learners show a larger effect size than advanced and elementary students. (c) child learners were better than adult learners in Learning L2 vocabulary. (d) Published studies generated larger effect size than doctoral dissertations. (e) employing “posters” for teaching L2 vocabulary items generated higher effect size than reading activities, CALL, and songs. (f) abstract words generated higher effect size than concrete ones. Possible explanations of the findings are discussed with regard to the similar meta-analyses in the field and directions for future research are proposed.

A Distributed Interior-Point KKT Solver for Multistage Stochastic Optimization

Multistage stochastic optimization leads to NLPs over scenario trees that become extremely large when many time stages or fine discretizations of the probability space are required. Interior-point methods are well suited for these problems if the arising huge, structured KKT systems can be solved efficiently, for instance, with a large scenario tree but a moderate number of variables per node. For this setting we develop a distributed implementation based on data parallelism in a depth-first distribution of the scenario tree over the processes. Our theoretical analysis predicts very low memory and communication overheads. Detailed computational experiments confirm this prediction and demonstrate the overall performance of the algorithm. We solve multistage stochastic quadratic programs with up to 400 × 106 variables and 8.59 × 109 KKT matrix entries or 136 × 106 variables and 12.6 × 109 entries on a compute cluster with 384 GB RAM. Data are available at https://doi.org/10.1287/ijoc.2017.0748 .

BoCluSt: Bootstrap Clustering Stability Algorithm for Community Detection.

The identification of modules or communities in sets of related variables is a key step in the analysis and modeling of biological systems. Procedures for this identification are usually designed to allow fast analyses of very large datasets and may produce suboptimal results when these sets are of a small to moderate size. This article introduces BoCluSt, a new, somewhat more computationally intensive, community detection procedure that is based on combining a clustering algorithm with a measure of stability under bootstrap resampling. Both computer simulation and analyses of experimental data showed that BoCluSt can outperform current procedures in the identification of multiple modules in data sets with a moderate number of variables. In addition, the procedure provides users with a null distribution of results to evaluate the support for the existence of community structure in the data. BoCluSt takes individual measures for a set of variables as input, and may be a valuable and robust exploratory tool of network analysis, as it provides 1) an estimation of the best partition of variables into modules, 2) a measure of the support for the existence of modular structures, and 3) an overall description of the whole structure, which may reveal hierarchical modular situations, in which modules are composed of smaller sub-modules.

Redundancy Gains in Communication Tasks: A Comparison of Auditory, Visual, and Redundant Auditory-Visual Information Presentation on NextGen Flight Decks

The redundant presentation of information in more than one sensory channel has traditionally been assumed to benefit performance. However, a recent meta-analysis suggests that redundancy gains may depend on task type and a number of moderator variables. The present study examined the effectiveness of visual, auditory, and redundant auditory-visual information presentation under high and low workload in the context of a mid-fidelity NextGen flight simulation with experienced airline pilots as participants. Overall, for two types of communication tasks – data link and ATIS – faster responses were observed for redundant displays, compared to vision and audition alone. No significant benefit of redundancy was found for accuracy due to a ceiling effect and workload did not mediate redundancy effects. The findings from this research add to the knowledge base in multimodal and redundant information processing and can inform modality choices in the design of displays for complex, data-rich domains.

Stability Investigations of Multivariable Regression Models Derived from Low- and High-Dimensional Data

Multivariable regression models can link a potentially large number of variables to various kinds of outcomes, such as continuous, binary, or time-to-event endpoints. Selection of important variables and selection of the functional form for continuous covariates are key parts of building such models but are notoriously difficult due to several reasons. Caused by multicollinearity between predictors and a limited amount of information in the data, (in)stability can be a serious issue of models selected. For applications with a moderate number of variables, resampling-based techniques have been developed for diagnosing and improving multivariable regression models. Deriving models for high-dimensional molecular data has led to the need for adapting these techniques to settings where the number of variables is much larger than the number of observations. Three studies with a time-to-event outcome, of which one has high-dimensional data, are used to illustrate several techniques. Investigations at the covariate level and at the predictor level are seen to provide considerable insight into model stability and performance. While some areas are indicated where resampling techniques for model building still need further refinement, our case studies illustrate that these techniques can already be recommended for wider use.

The Effects of Physical Activity and Physical Fitness on Children's Achievement and Cognitive Outcomes

It is common knowledge that physical activity leads to numerous health and psychological benefits. However, the relationship between children's physical activity and academic achievement has been debated in the literature. Some studies have found strong, positive relationships between physical activity and cognitive outcomes, while other studies have reported small, negative associations. This study was a comprehensive, quantitative synthesis of the literature, using a total of 59 studies from 1947 to 2009 for analysis. Results indicated a significant and positive effect of physical activity on children's achievement and cognitive outcomes, with aerobic exercise having the greatest effect. A number of moderator variables were also found to play a significant role in this relationship. Findings are discussed in light of improving children's academic performance and changing school-based policy.

An active set strategy for solving optimization problems with up to 200,000,000 nonlinear constraints

Numerical test results are presented for solving smooth nonlinear programming problems with a large number of constraints, but a moderate number of variables. The active set method proceeds from a given bound for the maximum number of expected active constraints at an optimal solution, which must be less than the total number of constraints. A quadratic programming subproblem is generated with a reduced number of linear constraints from the so-called working set, which is internally changed from one iterate to the next. Only for active constraints, i.e., a certain subset of the working set, new gradient values must be computed. The line search is adapted to avoid too many active constraints which do not fit into the working set. The active set strategy is an extension of an algorithm described earlier by the author together with a rigorous convergence proof. Numerical results for some simple academic test problems show that nonlinear programs with up to 200,000,000 nonlinear constraints are efficiently solved on a standard PC.

Are road safety evaluation studies published in peer reviewed journals more valid than similar studies not published in peer reviewed journals?

The peer review system of scientific journals is commonly assumed to prevent seriously flawed research from getting published. This paper compares the quality of 44 road safety evaluation studies published in peer reviewed journals to the quality of 79 evalutation studies dealing with the same safety measures, but not published in peer reviewed journals, in terms of seven criteria of study validity. Studies were scored for validity in terms of 1. (1) sampling technique, 2. (2) total sample size, 3. (3) mean sample size for each result, 4. (4)specification of accident or injury severity, 5. (5) study design, 6. (6) number of confounding factors controlled and 7. (7) number of moderator variables specified. Confounding factors are all factors that distrurb the attribution of a causal relationship between the safety measure being evalutated and the observed changes in safety, moderator variables are all variables that influence the size of the effect of the safety measure. Very few statistically reliable differences in study validity were found between studies published in peer reviewed journals and studies not published in such journals. There was, at best, a weak tendency for studies published in peer reviewed journals to score higher for validity. An interaction was found between author affiliation and type of publication with respect to study validity. Studies published in peer reviewed journals by authors who were at a university scored highest for validity. For a number of reasons, this study must be regarded as exploratory and its results as indicative only. The study does, however, point to a line of research that might be worth pursuing in larger and more rigorous studies.

Effect of therapeutic drug monitoring services on the number of serum drug assays ordered for patients: a meta-analysis.

In order to quantify the impact of therapeutic drug monitoring (TDM) services on the average number of serum drug assays (SDAs) ordered for patients, a meta-analysis was conducted on the data collected from 16 published and unpublished studies. Overall, TDM has no effect on the number of SDAs ordered per patient. However, there were a number of moderator variables that influenced the results. In studies utilizing control groups for comparison, TDM service-monitored patients were ordered fewer SDAs than nonmonitored patients. When data from the same institution were collected before and after implementation of a TDM service, there was an increase in the number of SDAs per patient. Aminoglycoside patients monitored by a TDM service were ordered more SDAs and nonaminoglycoside monitored patients were ordered fewer SDAs than nonmonitored patients. When the pharmacy department determined when the SDAs were to be collected and adjusted the dose, monitored patients were ordered fewer SDAs than nonmonitored patients. However, when the pharmacy department did not have control of these parameters, the TDM patients were ordered more SDAs than nonmonitored patients. Methodologic limitations of the studies and their possible effects on the results are discussed.