Needed: a ban on the significance test.
Part of a special section on the advantages and disadvantages of null-hypothesis significance testing in psychological research. The use of significance testing should be banned. Although most researchers believe that such testing has a 5-percent error rate, empirical studies indicate that the ave...
| Publicado en: | Psychological Science. Vol. 8; pp. 3 - 8 |
|---|---|
| Autor principal: | |
| Formato: | Artículo |
| Publicado: |
HWW Temp
January 1997
|
| Materias: | |
| Acceso en línea: | Ver este registro en EBSCOhost |
| fields | @attributes: recordID: 1 pdfLink: plink: https://search.ebscohost.com/login.aspx?direct=true&db=ssf&AN=512897948&site=ehost-live header: @attributes: shortDbName: ssf uiTerm: 512897948 longDbName: Social Sciences Full Text (H.W. Wilson) uiTag: AN controlInfo: bkinfo: jinfo: jid: 09567976 DA2U jtl: Psychological Science. issn: 09567976 maglogo: N pubinfo: dt: January 1997 vid: 8 pid: 71778 pub: HWW Temp artinfo: ui: 512897948 10.1111/j.1467-9280.1997.tb00534.x ppf: 3 ppct: 5 formats: tig: atl: Needed: a ban on the significance test. aug: au: Hunter, John E. su: Statistical significance Psychological techniques Psychology -- Statistical methods sug: subj: Statistical significance Psychological techniques Psychology -- Statistical methods ab: Part of a special section on the advantages and disadvantages of null-hypothesis significance testing in psychological research. The use of significance testing should be banned. Although most researchers believe that such testing has a 5-percent error rate, empirical studies indicate that the average error rate across psychology is 60 percent. As a result, comprehensive reviews cite conflicting findings on almost every issue, and writers who base their reviews on interpreted study findings have a 100-percent error rate in making such conclusions. However, these high error rates can be reduced using Cohen's power analysis longtime program and Harris's new program that derives from the previously suggested three-tailed test concept. Moreover, meta-analyses and confidence intervals for measuring the potential sampling error can be used in multiple studies and single studies, respectively. pubtype: Periodical doctype: Article src: R language: English refInfo: copyright: @attributes: flag: N holdings: @attributes: islocal: N |
|---|