A5012577066- Advanced Statistical Methods and Models
- Statistical Methods in Clinical Trials
- Optimal Experimental Design Methods
- Statistical Methods and Bayesian Inference
- Behavioral and Psychological Studies
- Advanced Causal Inference Techniques
- Psychometric Methodologies and Testing
- Mental Health Research Topics
- Child and Adolescent Psychosocial and Emotional Development
- Statistical Methods and Inference
- Advanced Statistical Modeling Techniques
- Meta-analysis and systematic reviews
- Statistical Methods and Applications
- Early Childhood Education and Development
- Sensory Analysis and Statistical Methods
- Attachment and Relationship Dynamics
- Behavioral Health and Interventions
- Cognitive and developmental aspects of mathematical skills
- Cognitive Abilities and Testing
- Emotional Intelligence and Performance
- Neural and Behavioral Psychology Studies
- Educational and Psychological Assessments
- Forecasting Techniques and Applications
- Advanced Statistical Process Monitoring
- Statistics Education and Methodologies
University of Notre Dame
2013-2025
Veeco (United States)
2022
Imperial College London
2020
Christchurch Hospital
2020
National University of Singapore
2017
NHS Lothian
2016
University of New England
2010
Notre Dame of Dadiangas University
2001
Notre Dame University
1985
University of Houston
1979-1983
R. M. Baron and D. A. Kenny (1986; see record 1987-13085-001) provided clarion conceptual methodological guidelines for testing mediational models with cross-sectional data. Graduating from to longitudinal designs enables researchers make more rigorous inferences about the causal relations implied by such models. In this transition, misconceptions erroneous assumptions are norm. First, we describe some of questions that arise (and sometimes emerge) in tests We also provide a collection tips...
Most empirical tests of mediation utilize cross-sectional data despite the fact that consists causal processes unfold over time. The authors considered possibility longitudinal might occur under either two different models change: (a) an autoregressive model or (b) a random effects model. For both models, demonstrated approaches to typically generate substantially biased estimates parameters even ideal conditions when is complete. In where variable M completely mediates effect X on Y, direct...
Maxwell and Cole (2007) showed that cross-sectional approaches to mediation typically generate substantially biased estimates of longitudinal parameters in the special case complete mediation. However, their results did not apply more typical partial We extend previous work by showing substantial bias can also occur with In particular, analyses imply existence a indirect effect even when true is zero. Thus, variable found be strong mediator analysis may at all analysis. addition, we show...
Psychology has recently been viewed as facing a replication crisis because efforts to replicate past study findings frequently do not show the same result. Often, first showed statistically significant result but does not. Questions then arise about whether results were false positives, and correctly indicates that there is truly no effect after all. This article suggests these so-called failures may be at all, rather are of low statistical power in single studies, failure appreciate need...
Underpowered studies persist in the psychological literature. This article examines reasons for their persistence and effects on efforts to create a cumulative science. The "curse of multiplicities" plays central role presentation. Most psychologists realize that testing multiple hypotheses single study affects Type I error rate, but corresponding implications power have largely been ignored. presence hypothesis tests leads 3 different conceptualizations power. Implications these are...
Despite pleas from methodologists, researchers often continue to dichotomize continuous predictor variables. The primary argument against this practice has been that it underestimates the strength of relationships and reduces statistical power. Although is correct for involving a single predictor, different problem can arise when multiple predictors are involved. Specifically, dichotomizing 2 independent variables lead false significance. As result, typical justification using median split...
This article extends current discussion of how to disaggregate between-person and within-person effects with longitudinal data using multilevel models. Our main focus is on the 2 issues centering detrending. Conceptual analytical work demonstrates similarities differences among 3 approaches (no centering, grand-mean person-mean centering) relations various detrending detrending, X only, Y both Y). Two real analysis examples in psychology are provided illustrate results different methods for...
We review recent developments in structural equation modeling (SEM) since Fassinger’s (1987) introductory article. discuss issues critical to designing and evaluating SEM studies. Our includes technological such as new approaches from hypothesis testing determining statistical power assessing model fit. Moreover, we innovations applying different research contexts designs (e.g., experimental longitudinal interactions among latent variables). Finally, procedures for redressing common problems...
Introduction to Multivariate Analysis of Variance Omnibus MANOVA Tests Analyzing and Interpreting Significant MANOVAs Causal Models Underlying Analyses Complex Designs Overview Computer Programs for
True experimental designs (Designs 4, 5, and 6 of Campbell & Stanley, 1963) are thought to provide internally valid results. This paper describes five studies involving the evaluation various treat ment interventions identifies a source inter nal invalidity when self-report measures used in Pretest-Posttest manner. An alternative approach (Retrospective design) measuring change is suggested, data comparing its ac curacy with traditional design treatment effects presented. Finally,...
Designing Experiments and Analysing Data. By S. E. Maxwell H. D. Delaney. ISBN 0 534 10374 X. Wadsworth, Belmont, 1990. xvi + 902 pp. £19.95 (paperbound).
The sample size necessary to obtain a desired level of statistical power depends in part on the population value effect size, which is, by definition, unknown. A common approach sample-size planning uses from prior study as an estimate be detected future study. Although this strategy is intuitively appealing, effect-size estimates, taken at face value, are typically not accurate estimates because publication bias and uncertainty. We show that use often results underpowered studies, sometimes...
"We propose to change the default P-value threshold forstatistical significance for claims of new discoveries from 0.05 0.005."
Despite the development of procedures for calculating sample size as a function relevant effect parameters, rules thumb tend to persist in designs multiple regression studies. One explanation their persistence may be difficulty formulating reasonable priori value an detected. This article presents methods sizes from variety perspectives and also introduces new method based on exchangeability structure among predictor variables. No single is deemed superior, but rather examples show that...
The development of child and adolescent self‐concept was examined as a function the domain, social/developmental/educational transitions, gender. In two overlapping age cohorts public school students ( N s = 936 984), five dimensions were evaluated every 6 months in manner that spanned grades 3 through 11 (representing elementary, middle, high years). Domains included academic competence, physical appearance, behavioral conduct, social acceptance, sports competence. Structural equation...