Referees Often Miss Obvious Errors in Computer and Electronic Publications

Abstract

Misconduct is extensive and damaging. So-called science is prevalent. Articles resulting from so-called science are often cited in other publications. This can have damaging consequences for society and for science. The present work includes a scientometric study of 350 articles (published by the Association for Computing Machinery; Elsevier; The Institute of Electrical and Electronics Engineers, Inc.; John Wiley; Springer; Taylor & Francis; and World Scientific Publishing Co.). A lower bound of 85.4% articles are found to be incongruous. Authors cite inherently self-contradictory articles more than valid articles. Incorrect informational cascades ruin the literature's signal-to-noise ratio even for uncomplicated cases.

Keywords:

• citations
• computer science
• electronic engineering
• Impact Factor
• misconduct
• refereeing shortcomings

ACKNOWLEDGMENT

I acknowledge the Associate Editor David B. Resnik and referees for constructive criticism. I hope that they do not feel that I am guilty of plagiarism by incorporating feedback from them.

Notes

1. Unfortunately publishers’ search engines produce results which are completely irrelevant if the term is “C++” (such irrelevant articles happen to have the letter ‘C’ in an article (for example if the address is Canadian or if the article is about computers). Furthermore, one publisher's search engine found no articles when queried for the attributive adjective “object-oriented” but found articles (containing this attributive adjective!) when queried for the predicative adjective “object oriented.” Such search terms were adjusted to peculiarities of publishers’ search engines in order to find a sizable quantity of articles per publisher. Search engines’ results were culled for occurrences of “C++.” This could be automated for some cases, but for others “C++” was found manually because awkward typesetting obstructed search software (for example, Kwan and Li, 1999, p. 227).