| 331 | Thursday, March 18, 2010 |
|
| 8:45 AM-10:00 AM | Convention Center:110 |
| Research Consortium |
| “p < .05” Abuse, Consequences, and Alternatives |
| It is all too common today to see researchers make their “significance” claims or statements based on merely “p < .05,” a single piece of information from their statistical significance tests. This symposium calls for exposing the p value abuse in our research practices. After a brief historical review of statistical significance testing and its positive role in helping scientific decisions, its limitations and widely spread p value abuse phenomenon will be described in detail. More appropriate ways of using information generated from statistical significance tests will be introduced. Finally, related software will be introduced and illustrated. |
Keyword(s): assessment, measurement/evaluation, research
|
| Presider: Marco S. Boscolo, University of Illinois at Urbana-Champaign, Urbana, IL |
| Speakers: Weimo Zhu, University of Illinois - Urbana-Champaign, Urbana, IL; Elena Boiarskaia and Marco S. Boscolo, University of Illinois at Urbana-Champaign, Urbana, IL; and Youngsik Park, University of Illinois at Urbana-Champaign, Urbana, IL |
|
Speakers: |
| |
Sadly, the Earth Is Still Round “p < .05”
Weimo Zhu, University of Illinois - Urbana-Champaign, Urbana, IL |
| |
Determining “Significance” Using Effect Size and Meta-Analysis
Elena Boiarskaia, University of Illinois at Urbana-Champaign, Urbana, IL and Youngsik Park, University of Illinois at Urbana-Champaign, Urbana, IL |
| |
“Practical Significance:” That Is What We Should Care About
Marco S. Boscolo and Bhibha M. Das, University of Illinois at Urbana-Champaign, Urbana, IL |
|