The Meta-Analyst’s Checklist
Intro
- Research
domain - Define Constructs
- Research
question(s) to be addressed
- Confirm
popular opinion
- Controversy
in the literature
- Problems
with previous meta-analyses
- Theoretical
justification of choice of moderators
Data Collection
Sources
- Conference
proceedings/abstracts
- Dissertations
- Electronic
databases searched; include terms, keywords
- PychInfo
- PsychArticles
- ScienceDirect
- ERIC
- Web
of Science
- ABI
Inform
- Medline
- Dissertation
Abstracts
- Etc.
- Hand
searches of literature – a must include because you get stuff you don’t
get in electronic databases
- Listservs
- Other
sources of documents (e.g., key researchers/informants)
- Reference
sections of articles
- Unpublished
manuscripts
- Websites
of authors in the area
Eligibility:
Inclusion/Exclusion Criteria
- Measures
allowed
- Characteristics
of the measures (e.g., minimum number of items, reliability, etc.)
- E.g.,
NEO vs. big five generic
- Specific
measures (e.g,. facet vs. general satisfaction)
- Participants
(e.g., all male)
- Publications
(type included)
- Journal
(e.g., APA journal)
- Book
vs. article, etc.
- Published
vs. not vs. subsequently published
- Study
design (e.g., specific control groups, repeated measures, etc.)
- Time
frame
- Required
information
- Effect
size
- Moderator
- Sample
size
- Other
(e.g., biological verification)
Coding Studies
- Choice
of effect size
- Assumptions
(e.g., if only p value reported) for derivation
- Transformations
from one effect size to another
- Code-book
description
- Consensus
method
- Description
of raters
- Rater
training, N of raters for each judgment
- Rater
Agreement/reliability
- Alpha/intraclass correlation
- Kappa
- Percent
agreement
- Test-retest
Data Analysis
Focal variables
- Bias
reduction (if needed; available for r
and d)
- Choice
of effect size
- Choice of Fixed-, Random- and
Mixed-effects
- Homogeneity
analysis (Q test)
- Imputed
data (e.g., substitute mean for missing reliability)
- Outliers
- Weights
- Inverse
variance
- Reliability
- Range
Restriction
- Sample-size
- Unit?
Moderators
- Categorical
- Continuous
- Homogeneity
analysis after modeling moderators
- Choice of models (fixed vs mixed)
Results
- (Weighted)
Mean ES
- CI
for mean
- Descriptive
stats (k, Ns, Means, SDs, etc.)
- Descriptions
of the measures (e.g., names, reliability estimates of raw variables)
- Optional
- Failsafe N/ filedrawer – report choice of min
for ES
- Graphic
representation of ES distribution(s) [some choices]
- Boxplot
- Funnel
plot
- Dot
(forest) plot
- QQ
plot
- Scatterplot
- Stem-leaf
- RE
Variance Component
- Results
of moderator analysis
- Between
& within values of Q (homogeneity)
- Report
means
- Report
regression coefficients/data
- Table
of description of the distributions by levels
- Sensitivity
of results to choices, e.g., outliers
- Variance
of ES
Threats to Validity
- Apples
& Oranges
- Dependent
Effect Sizes
- File
drawer
- Methodological
Rigor - GIGO
- Power/small
k (small N)
- Unanswerable
questions (e.g., correlational ESs, causal question)
Discussion
- Discussion
of credible threats / study limits
- Future
research e.g., other moderators, what are gaps based on what is known
- Implications consistent with data and
analysis
- Meaning
of results in terms of study (e.g., BESD)
- Proper
interpretation of the effect size (d, r, OR, etc.)
- Results
clearly linked to questions