1 is the only acceptable acceptance ratio

… which is of course a bit of a polemic title, let me try to motivate:

• I do not understand how acceptance ratio can serve as quality measure.
• Low acceptance ratios waste community resources and suggest exclusivity based attitudes towards quality.
• See this post on numerical evaluation criteria, including h-index.

How can acceptance ratio serve as measure of quality of a conference

Acceptance ratio, i.e., the fraction of total number of submitted papers over the number of accepted papers, is considered as important quality indicator of a conference.
I do not understand (as in I do not have the knowledge) how or why and believe there is a number of issues:

• It is very easy to manipulate: e.g. submit arbitrary number of random papers, who actually checks that the numbers given are true?
• It is a relative measure, it
  • does not say of how high a quality a paper is
  • but only how many other papers were kept out
  • based on an “us vs. them” understanding of community
• It is based on influences outside the control of authors, varying over time
  • e.g., how many papers were submitted, possibly on certain topics, in a year
  • the same paper would be accepted at different acceptance ratios, in e.g. different years of the same conference
• It may force rejection of strong papers
  • “so we have received papers from Einstein, Higgs, Planck, Bor, etc, but we can accept only 10%, who do we reject?”
  • “Clearly this paper will solve the climate crisis and can bring world peace – but if I accept more than 10% I or the conference look bad”

Acceptance ratio is first a function of which and how many papers were submitted in given year, based on how many papers were kept out. What does this say about quality of papers, I do not see how it can be used to compare or evaluate.

Focusing on acceptance ratio hurts the community

Apart from not being the meaningful quality measure it is often used, I believe it is even bad for the community. As stated above, it is a measure of many papers were kept out (in a given year, depending on other papers etc)

• community understanding
  • it hurts the community and ultimately society by possibly rejecting very good work or delaying its publication
  • it does not evaluate quality by itself, but is based on a notion of how many papers were left out
    • it suggests a community of “us”, we who know the game and take decisions, vs. “them” those we keep out
    • it provides an incentive to increase the gap between “us” and “them”, as rejecting many papers. “them”, increases the perceived quality of accepted papers (low acceptance ratio), “us”
      • those who define the rules profit from high rejection numbers
  • it suggests a community based on exclusivity, serves those inside rather than the whole community
• community resources
  • writing papers, submitting, reviewing, evaluation etc requires substantial resources from those involved, authors, reviewers, technical program committee, etc, i.e., the community
  • a low acceptance ratio means, put a bit polemically, that the conference managed to lure many authors to submitting papers, which will not get accepted, wasting substantial community resources in the process
    • “you all work hard, mostly in vain, so the conference can look good”
  • junior researchers and TPC member are disproportionately affected
• acceptance ratio one
  • would mean
    • no efforts wasting on rejected papers
    • community resources only used where they bring benefit to the community
    • inclusiveness as basis for community, “us” could become the entire community

Acceptance Ratio One is practically not possible, but I believe it can serve as guideline for different quality measures and community attitude.

A fictional story

… illustrating some of the points.
Three seasoned TPC chairs of conferences discuss in a panel, comparing the quality of their conferences:

• A: “We accepted all papers we found to be of good quality.”
• B: “Nice and commendable, but what was your acceptance ratio?”
• A: “50 %”
• audience gasps, B and C shake their heads “that cannot be a good conference”
• B: “We never had higher acceptance ratio than 10%.”
• audience looks impressed
• C smiles condescendingly: “Decent conference, but you might still have accepted too many papers. Only our conference can claim to be perfect pillar of quality, never having accepted too many papers: we never accepted a single paper, we are the best conference based on acceptance ratio!”
• gasps and admiration, TPC chair, scattered audience members: “but….”