MetaAnalysis

1
Chapter 11

• Meta-Analysis

2

• Supposing a large number research has been done
  on a particular subject.
• Results are inconclusive some say yes, some say
  no.

3

• If we examine the available research carefully,
  we find that the various studies differ in terms
  of the way to research was conducted sample
  (N), procedure, age, gender

4
Meta-analysis

• Definition quantitative means of reanalyzing the
  results from a large number of research studies
  in an attempt to draw conclusions which are
  supported by many research studies.

5

• Quantitative means of reanalyzing the results
  from a large number of research studies in an
  attempt to synthesize findings.
• More than merely a review of related literature!

6

• With meta-analysis, results from available
  research studies are reanalyzed.
• Allowances are made for the variables that need
  to be controlled because they may influence
  findings in a research study.

7

• Studies that are too small or marginally
  acceptable procedures may not be used.

8

• Meta-analysis has definite procedures with
  systematic steps helps the researcher in
  drawing defensible conclusion.
• A large of research can be handled in an economic
  manner wont overwhelm the reader.

9
Meta-Analysis

• In meta-analysis, each research study contributes
  a data point to the subsequent analysis, much
  like an individual participant in a descriptive
  or experimental research study.

10
Effect Size

• A basic statistic used in meta-analysis, which
  converts results from different studies to a
  common metric so that comparisons can be made.
• Used to estimate meaningfulness of an outcome
  (i.e., practical significance).

11
Example Size Formula

• ES (Me Mc)/Sc
• Formula for estimating ES for difference between
  experimental and control group.
• Where Me is the mean of the experimental group,
  Mc is the mean of the control group, and Sc is
  the standard deviation of the control group.

12
Procedures in Meta-analysis

• Compile references must be a substantial number
  of research studies available on a topic!
• Requires means, standard deviations,
  correlations, and such to be published.

13

• Determine your criteria make a basic list of
  things that must be in the study before you
  review them.

14

• For example, if the area is changing rapidly, you
  might not review any research study that is more
  that 10 years old.

15

• Review each study
• Good notes concerning
• Basic information,
• Number of participants,
• Number of treatment used,
• Length of time,

16

• Quality of the procedures,
• Findings,
• Information needed to calculate effect size
• Identify moderator variables (?)

17

• Moderator variables factors which influence the
  relationship between two variables or effect
  size.
• They have to be identified and coded.

18

• Its important to get the necessary information
  and apply the same standards to all research
  papers.
• The whole idea of meta-analysis is to remove
  subjectivity as much as possible

19

• Decide which studies youll use what study will
  be used in the meta-analysis, and decide whether
  moderator variables will be used.

20

• Do the meta-analysis calculate an effect side
  for each research study used, and generate
  summary statistics for effect sizes.
• Report the results.

21
Computer Program

• When youre planning on incorporation a large
  of articles, useful to do organizing and
  computing.
• May be more accurate!
• Number of software, e.g. DSTAT

22
DSTAT

• Literature review tactics
• Conducting a meta-analysis review
• General program operations and installtion
• Calculation of effect size
• Management of effect size
• Analysis of effect size

23
Criticism of Meta-Analysis

• Not the ultimate answer!
• Does not differentiate in quality of studies.
• Combines unlike studies with too much variability.