Acceptance rate versus impact

Should you attend the most selective school? Maybe not:

Students who attended more selective colleges do not earn more than other students who were accepted and rejected by comparable schools but attended less selective colleges. (Dale and Krueger, Estimating the payoff to attending a more selective college, 1999).

Should you present papers in the conference with the lowest acceptance rate? Looking at this plot, there seems to be little correlation between acceptance rate and impact factor:

acceptance rate versus impact factor

(Source: Sylvain Hallé’s blog.)

Conclusion: The best schools or the best conferences may not be those with low acceptance rates.

Published by

Daniel Lemire

A computer science professor at the University of Quebec (TELUQ).

6 thoughts on “Acceptance rate versus impact”

  1. Interesting. However, A BIG assumption in your inference is that “good conference” is related to “impact factor”. There are lots of reasons to doubt whether “impact factor” of a journal has any relationship at all with “quality of publication”.

  2. That’s likely because the earning power (or impact) isn’t in the school (conference).

    On the other hand, I imagine there might be some interesting observations to make regarding the whole distributions, rather than just the means.

  3. @Vellino I’m not making this assumption. I have stopped tracking the Impact Factor entirely after giving it a bit of thought.

    But here, it suits my purpose which is to question this concept that “low acceptance rates” are a good thing.

  4. Is there a bias here? I assume there are many more low-acceptance rate conferences than high-acceptance. Therefore, it makes sense that there are more papers which were accepted to low-acceptance conferences, and then gained high impact.

    The question should be different, I think. If you have a paper that was accepted to a high-acceptance conference, what are the chances it will gain high impact factor over time?

  5. I’ve often felt this is true, but I wonder how things stack up within specific areas (eg., theory versus theory or AI versus AI)…

  6. It’s clear that having low acceptance rates for the sake (or prestige) of having low acceptance rates is not a useful signal detection (detection of quality work) strategy. But what is a good signal detection strategy? And what contributes to a low acceptance rate, is it a high rejection bias or a low submission threshold? Teasing these apart might yield related matrics that are more meaningful.

Leave a Reply

Your email address will not be published. Required fields are marked *

To create code blocks or other preformatted text, indent by four spaces:

    This will be displayed in a monospaced font. The first four 
    spaces will be stripped off, but all other whitespace
    will be preserved.
    Markdown is turned off in code blocks:
     [This is not a link](

To create not a block, but an inline code span, use backticks:

Here is some inline `code`.

For more help see