{"id":933,"date":"2012-01-18T17:23:38","date_gmt":"2012-01-18T17:23:38","guid":{"rendered":"http:\/\/occamstypewriter.org\/trading-knowledge\/?p=933"},"modified":"2012-01-18T17:23:38","modified_gmt":"2012-01-18T17:23:38","slug":"err-ology","status":"publish","type":"post","link":"https:\/\/occamstypewriter.org\/trading-knowledge\/2012\/01\/18\/err-ology\/","title":{"rendered":"Err … ology"},"content":{"rendered":"

Publication bias is the tendency to report positive results differently from negative or inconclusive results, resulting in a bias in the overall literature (see Wikipedia article<\/a>\u00a0and this\u00a0tutorial at the Cochrane Collaboration<\/a>). Afficionados of evidence-based practice and meta-analysers of research worry that such bias makes it hard to accurately interpret the literature (though there is bias and bias<\/a>). \u00a0In the clinical field it was hoped that registration of clinical trials would improve the situation, creating more pressure to publish results of all trials. \u00a0Nature <\/em>reported\u00a0<\/a>in 2009 that<\/p>\n

“Fewer than half of published trials are adequately registered, and, on the other hand, fewer than half of registered trials are ever published in peer-reviewed journals”.<\/p><\/blockquote>\n

In a letter to Nature last year Bob O’Hara<\/a> pointed out that a number of journals do publish negative results, listing the following as titles devoted to or including negative results:<\/p>\n