Erschienen in:
06.10.2018 | COMMENTARY
Massive citations to misleading methods and research tools: Matthew effect, quotation error and citation copying
verfasst von:
John P. A. Ioannidis
Erschienen in:
European Journal of Epidemiology
|
Ausgabe 11/2018
Einloggen, um Zugang zu erhalten
Excerpt
Research methods and tools comprise a lion’s share among the most cited papers across science [
1]. Methodological tools are essential to make discoveries, assess them, organize our knowledge, and understand which information is valid and useful. Many methods and research tools are proposed, but few become widely utilized. These are not always the best. For example, null-hypothesis significance testing with reporting of p-values is embedded in millions of papers [
2], despite being a poor inferential method for most [
3]. The factors that shape which methodological paper gets widely cited are poorly known. However, perhaps methods that are simple and easy to use (or misuse), and those that address major needs are more prone to become popular. Conversely, esoteric and convoluted tools, those that are not readily practicable, and those that have relevance only to rare circumstances are unlikely to become citation classics. …