Do citation counts measure research quality?
No, citations measure attention, with a lag and heavy field effects, and are an error-prone proxy for merit.
Citation counts capture how much a paper is noticed, which is not the same as whether it is correct or well-executed. Counts are shaped by field size, time since publication, review versus primary research, self-citation, and even controversy; a wide review of citing behaviour found it reflects many motives beyond endorsing quality (Bornmann & Daniel 2008). Tested directly as a quality signal, citations perform poorly, and evaluators tend to over-rate highly cited work (Eyre-Walker & Stoletzki 2013). The signal is also degrading: fabricated and paper-mill references are entering the literature, so a high count can now reflect manipulation rather than merit (Topaz et al. 2026). Citations are a useful map of influence, but a weak measure of soundness - a separate question, answered by reading the work.
Sources
Frequently asked
Don't good papers get cited more?
Often, but unevenly and slowly — many strong papers are under-cited and flawed ones over-cited (Eyre-Walker & Stoletzki 2013).
Is the h-index a better measure?
It inherits the same problems, since it’s built from citation counts (Bornmann & Daniel 2008).
Related
Last updated