Referring to the file drawer problem and data falsification, the
president of the Association for Psychological Science recently wrote: "These
and related factors should tend to inflate 'false positives' (aka Type I
errors), which leads inexorably to the pessimistic conclusion that some unknown
(but considerably higher than .05) proportion of our field's published effects
are not true effects."
Am I misreading this, or does he imply (incorrectly) that use of the
.05 criterion of statistical significance would be expected to result in 5% of
published research findings being Type I errors?
---
You are currently subscribed to tips as: [email protected].
To unsubscribe click here:
http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=17069
or send a blank email to
leave-17069-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
