Upload
bart
View
23
Download
0
Tags:
Embed Size (px)
DESCRIPTION
RStats Statistics and Research Camp 2014. Meta-Analysis Session 4. Melissa Maier, Ph.D. Assistant Professor Communication. Rationale. Narrative Review Meta-Analysis: Mathematical Reduce Type II error Correct statistical artifacts Test possible moderator variables - PowerPoint PPT Presentation
Citation preview
Slide 1
RStats Statistics and Research Camp 2014
Meta-AnalysisSession 4
Melissa Maier, Ph.D.Assistant Professor
Communication
Slide 2
Rationale
• Narrative Review• Meta-Analysis:–Mathematical• Reduce Type II error• Correct statistical artifacts• Test possible moderator variables• Evaluate theoretical arguments
– Practical
Slide 3
Method
1. Construct database of all relevant research2. Analyze articles to determine (and correct):
1. Sample size2. Effect size3. Moderators
3. Calculate average effect size4. Test for homogeneity5. If heterogenous, test for moderators and
outliers
Slide 4
Calculate Ave. Effect
Observed Effect (A)
Sample Size (B)
(A*B)
Study 1 .87 162 140.94Study 2 .86 151 129.86Study 3 .83 22 18.26Study 4 .77 121 93.17Study 5 .58 113 65.54Sum 569 447.77
d(ave) = ∑ (Observed effect * sample size) = 569 = 1.27∑(Sample Size) 447.77
Are female same-sex relationships more intimate than male same-sex relationships?
Slide 5
Test for homogeneity
d d – ave d (d – ave d)2 (d – ave d)2(N-k)
Study 1 .87 (.87 – 1.27) = -.40 .16 (.16) (162-5) = 25.12Study 2 .86 -.41 .17 24.82Study 3 .83 -.44 .19 3.23Study 4 .77 -.50 .25 29Study 5 .58 -.69 .48 52.84Sum 135.01
Testing for homogeneity
Χ2 = (d – ave d)2(N-k)
Slide 6
Write-up• Justification for review• Methodology
– Describe search methods– Code possible moderators (or model)– Describe statistical procedures
• Results– Average effect– Number of studies, k– Overall combined sample size, N– Measure of variability– Evaluation of homogeneity– Measure of significance of average effect
• Discussion
Slide 7
Limitations of Meta-Analysis
• Contextual restrictions• Unequal value of claims• Ethnographic trap• Multiplicity of interpretation• Misapplication of the level of
analysis
Slide 8
Supplemental ResourcesBurrell, N.A., Allen, M., Gayle, B.M., & Preiss, R.W. (Eds). (2014). Managing
interpersonal conflict: Advances through meta-analysis. New York: Routledge.
Hunter, J.E., & Schmidt, F.L. (1990). Methods of meta-analysis: Correcting error and bias in research findings. Newbury Park, CA: Sage.
Hunter, J.E., & Schmidt, F.L. (2002). Methods of meta-analysis: Correcting error and bias in research findings (2nd ed.). Thousand Oaks, CA: Sage.
Hunter, J.E., Schmidt, F.L., & Jackson, G.B. (1982). Meta-analysis: Cumulating research findings across studies. London: Sage.
Lipsey, M.W., & Wilson, D.B. (2001). Practical meta-analysis. Thousand Oaks, CA: Sage.
Preiss, R., & Allen, M. (1995). Understanding and using meta-analysis. Evaluation & the Health Professions, 18, 315-335.
Preiss, R., & Allen, M. (2001). Understanding and using meta-analysis. In R. Preiss, B. Gayle, N. Burrell, M. Allen, & J. Bryant (Eds.), Mass media effects research: Advances through meta-analysis (pp. 15-30). Mahwah, NJ: Lawrence Erlbaum.The Meta-Analysis Calculator:
http://www.lyonsmorris.com/ma1/