1 / 29

Meta-Analysis

Meta-Analysis. A statistical method for cumulating studies. Why is Meta-Analysis Better Than Traditional Reviews?. Correlation Between Credit Rating and Job Performance. Correlation Between Credit Rating and Job Performance. Meta-Analysis Steps. Obtain relevant studies

Download Presentation

Meta-Analysis

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Meta-Analysis A statistical method for cumulating studies

  2. Why is Meta-Analysis Better Than Traditional Reviews?

  3. Correlation Between Credit Rating and Job Performance

  4. Correlation Between Credit Rating and Job Performance

  5. Meta-Analysis Steps • Obtain relevant studies • Convert test statistics into effect sizes • Compute mean effect size • Correct effect sizes for sources of error • Determine if effect size is significant • Determine if effect can be generalized or if there are moderators

  6. Finding Studies • Establish time frame for studies • Sources • Journals • Dissertations • Theses • Technical reports • Conference presentations • File cabinet data

  7. Finding StudiesMethods • Search Engines • Infotrac • PsychInfo • Lexis-Nexis • Factiva • World Cat • Internet • Bibliographies from studies • Phone calls • List serve calls for help

  8. Finding StudiesDeciding Which Studies to Use • Must be empirical • Must have the appropriate statistic to convert to an ‘r’ or a ‘d’ • Must have complete set of information • Must be accurate

  9. Converting Test Statistics into Effect Sizes • Two common effect sizes • Correlation (r) • Difference (d) • Conversion Types • Directly using means (Mexp – Mcontrol) ÷ SDoverall • Formulas to convert t, F, X2, r, and d

  10. Let’s Practice!

  11. Let’s Practice!

  12. Cumulating Effect Sizes

  13. Let’s Practice!

  14. Correcting Correlations for Common Sources of Error • Test unreliability • Criterion unreliability • Restriction of range

  15. Validity = .30 Test reliability = .90 Criterion reliability = .50

  16. Example 1 Validity = .10 Test reliability = .75 Criterion reliability = .80 Example 2 Validity = .20 Test reliability = .95 Criterion reliability = .90 Let’s Practice!

  17. Interpreting Meta-Analysis Results

  18. Scope of the Meta-Analysis • Number of Studies • Number of studies • Number of actual data points (k) • Allowing more than one data point per study • Sample size • Number of participants (n)

  19. The Effect Size • Correlations • Observed mean correlation (r) • Corrected correlation • Criterion reliability • Predictor reliability • Range restriction • Difference scores • d • Z

  20. Statistical Significance • Confidence Intervals • Uncorrected correlation • Significant if interval does not contain zero • Common intervals are 95%, 90%, and 80% • Credibility Intervals • Corrected correlation

  21. Can We Generalize Results? • Size of confidence and credibility interval • 75% sampling error rule • Statistical tests (all use chi-square distributions) • Chi-square (X2) • Homogeneity Test (HT) • QW

  22. Interpreting Meta-Analysis Results

  23. Interpreting Meta-Analysis Results

  24. Roth, BeVier, Switzer, & Shippmann (1996)

  25. Roth, BeVier, Switzer, & Shippmann (1996)

  26. Stem and Leaf Diagrams

  27. Let’s Practice!

More Related