Meta-analysis

1
Meta-analysis

• Overview

www.psychwiki.com/wiki/Meta-analysis_club_Summer20
08
2
Definition

• Definition
• A meta-analysis statistically combines the
  results of several studies that address a shared
  research hypotheses.
• A study collects data from individual subjects
  (such as 100 subjects 100 data
  points)
• A meta-analysis collects data from individual
  studies
• (such as 100 studies 100 data points)

3
Steps

• 1. Defines your hypothesis
• 2. Locate Studies
• 3. Find effect size for each study
• 4. Average the effect sizes together
• 5. If you want, you can analyze moderators
• 6. Write the manuscript

4
Why do a meta-analysis?

• Easy
• Steps are simple, there is software to calculate
  everything
• Cost-effective
• Since you have already read a bunch of articles
  to write a paper, not much more work to
  synthesize them together
• Best type of article
• Most highly cited type of article.
• Advantages of both qualitative and quantitative
  research
• Truly answers research questions within the
  literature (compared to single studies which
  cant truly generalize)

5
Step 2 (again)

• Locate studies
• Techniques - database searches, ancestry
  approach, descendancy approach, hand searching,
  invisible college
• Doesnt have to be comprehensive (fail-safe n)
  but needs to be close to comprehensive
• (create excel file example)
• so that you can Search it, copy/paste into
  References

6
Step 3 (again)

• Find effect size for each study
• (1) Decide convert into r or d
• (2) Download es_calculator.zip from
  http//mason.gmu.edu/dwilsonb/ma.html
• (3) Use es_calculator to calculate ES
• (4) Record in excel file, separate row for each
  study
• (excel file example)
  need to input sample
  size and ES

7
Step 4 (again)

• Average the effect sizes together
• Conceptually
• First, weight them by sample size/inverse
  variance
• Second, sum them together
• Third, divide by sum of total sample size.
• In Practice
• You calculate inverse variance
• Then use Macro to do the rest
• Download macro for SPSS http//mason.gmu.edu/dwi
  lsonb/ma.html

8
Step 5 (again)

• If you want, you can analyze moderators
• Conceptually
• Can test categorical moderators (categories like
  college student versus actual juror) similar to
  ANOVA
• Can test continuous moderators (such as length of
  stimulus) similar to Regression
• In Practice
• Use macros downloaded from http//mason.gmu.edu/d
  wilsonb/ma.html
• Macros exist for ANOVA Regression

9
Some cool things about meta-analyses
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)