Preprint Open Access

Estimation statistics should replace significance testing

Claridge-Chang, Adam; Assam, Pryseley

Dublin Core Export

<?xml version='1.0' encoding='utf-8'?>
<oai_dc:dc xmlns:dc="" xmlns:oai_dc="" xmlns:xsi="" xsi:schemaLocation="">
  <dc:creator>Claridge-Chang, Adam</dc:creator>
  <dc:creator>Assam, Pryseley</dc:creator>
  <dc:description>In place of significance testing, a preferred statistical methodology is accessible with modest re-training. However, an obstacle to the adoption of this alternative is a basic branding problem: it does not have a widely-used name. We suggest the most appropriate name for this superior approach is ‘estimation statistics,’ a term describing the methods that focus on the estimation of effect sizes (point estimates) and their confidence intervals (precision estimates). Estimation statistics offers several key benefits. 

This letter was previously published In Nature Methods at</dc:description>
  <dc:source>Nature Methods 13 108-109</dc:source>
  <dc:subject>estimation statistics</dc:subject>
  <dc:title>Estimation statistics should replace significance testing</dc:title>
All versions This version
Views 1,0621,065
Downloads 705705
Data volume 46.5 MB46.5 MB
Unique views 982985
Unique downloads 658658


Cite as