The New Statistics: Why and How

@article{cumming2014newstatistics,
author = {Geoff Cumming},
title ={The New Statistics: Why and How},
journal = {Psychological Science},
volume = {25},
number = {1},
pages = {7-29},
year = {2014},
doi = {10.1177/0956797613504966},
note ={PMID: 24220629}
}

The problem:

Research Integrity

  • Published research is a biased selection of all research;
  • data analysis and reporting are often selective and biased; and
  • in many research fields, studies are rarely replicated, so false conclusions persist.

a decision to report research […] must be independent of the results.

Agree, but…

The best way to ensure this is to make a commitment to report research in advance of conducting it

Not sure I agree.

No matter how intriguing, however, the results of such pilot work rarely deserve even a brief mention in a report.

Strongly disagree.

PHONEME TRANSPOSITION AND TEMPORAL ENCODING IN HUMAN SPEECH RECOGNITION - example pre-registration that seems relevant to the lab.