5 ms·
^ This. Peer-review is not a magical panacea that guarantees anything. The way to judge the impact of a scientific paper is through its citation count, perhaps
by arcanus 10y ago
^ This. Peer-review is not a magical panacea that guarantees anything.
The way to judge the impact of a scientific paper is through its citation count, perhaps normalized by a typical count for that particular field.
- kardos 10y agoUntil the sham articles start citing each other to game that approach
- JorgeGT 10y agoOr the classic... "Reviewer #1: Nice paper, but seems authors have missed this recently published work by TotallyNotReviewer1 et al. which could add..."
- visarga 10y agoSay you a are researcher leading a group. Then your name is going to appear on all papers published by your group (mentioned last). I don't think it should count the same as being the first mentioned author. Simple citation count is not good enough. Another situation: say you are the author of a dataset. Then all papers using your dataset are going to mention you, but such mentions are qualitatively different. They should not be counted in the same category.
- arcanus 10y agoConversely, when scanning a table of contents, I look at the last author first, because that tells me what group produced the work.
- detaro 10y agoImpact and correctness are two very different things. Cool results often get cited quite a bit, without much checking if the result actually is valid/can be repeated. (but yes, peer review doesn't solve this either, especially not in the current form)
- braveo 10y agoyep: see the current problems some of the social sciences are having with reproducibility.
- grigjd3 10y agoI know of research groups that intentionally cite other members of the same group to get citation numbers up. Sure, some of it is legit, but some is clearly not - particularly when actively ignoring relevant work from other groups.