Building a Better Bibliography
Posted: August 21, 2012 | Author: A. R. Hafdahl | Filed under: Uncategorized | Tags: bibliography, meta-analysis, methodology, research synthesis | Leave a commentMy previous blog post was more than two months ago. Since then I’ve nearly completed Part 6 in my 7-part overview of meta-analysis, about various types of problems that can arise. It’s long-ish and covers several pertinent issues, such as aberrant cases (e.g., outliers), sources of bias (e.g., reporting biases, artifacts), threats to validity, and sensitivity analyses. I’d like to post that before September.
Meanwhile, I invite you to test-drive the project that’s consumed much of my time during the past couple of months: building a CiteULike library on methodology for research synthesis (Meth4ReSyn). Read the rest of this entry »
Sneak Preview 2: Outliers, Metric Transformation, and ES Distribution
Posted: May 31, 2012 | Author: A. R. Hafdahl | Filed under: Sneak Preview | Tags: assumption violation, between-studies variance component, conditional variance, correlation, effect size, fixed effect, heterogeneity, interval estimation, meta-analysis, outlier, random effect, substantive application | Leave a commentMy previous three posts on fitting models to effect sizes (ESs)—Parts 5a, 5b, and 5c—were the core of my seven-part overview of meta-analysis. With only two posts remaining in the overview, I’ll pause again to describe three more methodological issues I plan to discuss: potential outliers, transforming ES metrics, and the distribution of ES parameters. As in my first sneak preview—about degraded ESs and tricky conditional variances (CVs)—I’ll keep these “teaser” descriptions fairly short, mainly to pique your interest; each issue deserves at least one dedicated post with more detail.
Read the rest of this entry »
Overview of Meta-Analysis, Part 5c (of 7): Primary Meta-Analyses (cont.)
Posted: May 13, 2012 | Author: A. R. Hafdahl | Filed under: Overview of Meta-Analysis | Tags: Bayesian analysis, between-studies variance component, dependence, fixed effect, heterogeneity, interval estimation, meta-analysis, meta-regression, model comparison, moderator, multivariate effect size, random effect, significance testing, standardized mean difference | 1 CommentThis is the last of three posts in Part 5 of my overview of meta-analysis. In Part 5a I described six conventional meta-analytic models for effect-size (ES) estimates, and in Part 5b I described estimation and inference for two of those models without covariates. In this post I’ll extend the methods of Part 5b to two models with covariates and comment on extensions and other variants of these models and procedures, to hint at the wide variety of situations that arise in meta-analysis. In Parts 6 and 7 of the overview, I’ll address follow-up procedures and ways to report results, respectively.
Read the rest of this entry »
Overview of Meta-Analysis, Part 5b (of 7): Primary Meta-Analyses (cont.)
Posted: April 30, 2012 | Author: A. R. Hafdahl | Filed under: Overview of Meta-Analysis | Tags: between-studies variance component, conditional variance, fixed effect, heterogeneity, interval estimation, math notation, meta-analysis, meta-regression, random effect, significance testing, standardized mean difference | 2 CommentsThis is the second of three posts in Part 5 of my overview of meta-analysis. In Part 5a I described six conventional models for meta-analysis, each of which combines within-study and between-studies models. In this second post I first comment on nested models then describe estimation and inference for two models without covariates—procedures for fitting these models to effect-size (ES) estimates and quantifying uncertainty about their focal (hyper)parameters. In the third post, Part 5c, I’ll do the same for two models with covariates and also comment on extensions and variants of these models and procedures.
Read the rest of this entry »
Overview of Meta-Analysis, Part 5a (of 7): Primary Meta-Analyses
Posted: April 12, 2012 | Author: A. R. Hafdahl | Filed under: Overview of Meta-Analysis | Tags: between-studies variance component, categorical data, conditional variance, effect size, fixed effect, heterogeneity, meta-analysis, meta-regression, moderator, multilevel model, random effect | 2 CommentsThe previous four parts of this seven-part overview of meta-analysis focused on obtaining data and preparing them for the central task addressed in this fifth part: meta-analyzing effect-size (ES) estimates, which I’ll cover in three subparts focused on meta-analytic models (Part 5a) and procedures for fitting them to ESs (Parts 5b and 5c). In the last two parts (6 and 7) I’ll address follow-up techniques to assess potential problems with these primary analyses, as well as useful ways to report these analyses’ results. (Topics for all seven parts of this overview are listed in Part 1.)
Read the rest of this entry »
Sneak Preview: Degraded Effect Sizes and Tricky Conditional Variances
Posted: April 2, 2012 | Author: A. R. Hafdahl | Filed under: Sneak Preview | Tags: Bayesian analysis, conditional variance, correlation, effect size, meta-analysis, missing data, resampling, significance testing, standardized mean difference, vote counting | Leave a commentMy post on data exploration more than half completed my seven-part overview of meta-analysis. As a diversion while I write Part 5, let’s consider two of several methodological issues I plan to discuss in this blog: degraded effect sizes (ESs) and tricky conditional variances (CVs). My main aim here is to pique your interest in future posts by offering a glimpse at ways to manage selected challenges that routine meta-analytic techniques don’t address. These “teaser” descriptions will be quite superficial. I plan to elaborate on each of these challenges—as well as many others—after laying a foundation in my seven-part overview.
Read the rest of this entry »
Overview of Meta-Analysis, Part 4 (of 7): Data Exploration
Posted: March 28, 2012 | Author: A. R. Hafdahl | Filed under: Overview of Meta-Analysis | Tags: coding, data management, effect size, meta-analysis, missing data, moderator, outlier, reporting guidance, sample size, significance testing | 1 CommentThis seven-part overview’s first three parts focused on collecting data used in meta-analyses: estimates of effect size (ES), sample sizes or conditional variances (CVs) to quantify ES sampling error or (im)precision, and ES features. The overview’s subsequent four parts address analyzing these data and presenting results. In this fourth part I begin by describing preliminary analyses that can help identify errors and issues to attend to in primary analyses. (Part 1 of this overview lists the topics for all seven parts.)
Read the rest of this entry »
Overview of Meta-Analysis, Part 3 (of 7): Effect-Size Features
Posted: March 19, 2012 | Author: A. R. Hafdahl | Filed under: Overview of Meta-Analysis | Tags: coding, effect size, inclusion/exclusion, individual participant data, meta-analysis, missing data, moderator, primary-study quality | Leave a commentIn Part 2 of this seven-part overview, I described obtaining the sample size(s) or conditional variance (CV) associated with an effect-size (ES) estimate to quantify this estimate’s sampling error or (im)precision. Here in Part 3 I’ll address coding features linked to ESs. Whereas this overview’s first three parts focus on collecting data used in a research synthesis, its subsequent four parts will address meta-analyzing these data and presenting results. (Part 1 lists the topics for all seven parts.)
Read the rest of this entry »
Overview of Meta-Analysis, Part 2 (of 7): Sampling Error
Posted: March 11, 2012 | Author: A. R. Hafdahl | Filed under: Overview of Meta-Analysis | Tags: binary outcome, conditional variance, correlation, dependence, effect size, heterogeneity, meta-analysis, missing data, multivariate effect size, primary-study design, sample size, standardized mean difference | Leave a commentIn Part 1 of this seven-part overview of meta-analysis, I introduced Conn, Hafdahl, Cooper, Brown, and Lusk’s (2009) quantitative review of workplace exercise interventions and discussed extracting effect-size (ES) estimates. Building on that material, in this second part I’ll address obtaining info about an ES’s sampling error, which plays a critical role in most modern meta-analytic methods. (Part 1 of this overview lists topics in the subsequent five posts.)
Read the rest of this entry »
Overview of Meta-Analysis, Part 1 (of 7): Effect Sizes
Posted: February 27, 2012 | Author: A. R. Hafdahl | Filed under: Overview of Meta-Analysis | Tags: binary outcome, correlation, dependence, effect size, meta-analysis, missing data, multivariate effect size, standardized mean difference, substantive application | Leave a commentThis post is the first in a seven-part overview of common meta-analytic tasks. In this first part I’ll introduce a real-world substantive application of meta-analysis and address estimating effect sizes (ESs). Subsequent parts will focus on the following topics:
- Part 2: obtaining information about ES sampling error
- Part 3: collecting features of ESs
- Part 4: exploring data
- Part 5: fitting meta-analytic models to ESs (subparts 5a, 5b, and 5c)
- Part 6: checking for potential problems
- Part 7: expressing results informatively
Meta-Analysis Sandwich 