journals - When and how often is an associate editor not starting review process?

Very simple question. Reading some Q&A here and experience from my own papers and reviews I'm actually wondering when and how often the associate editor managing a possible review process of a submitted paper declines to start the process.

Maybe for not reaching formal standards of the journal, or wrong scope. If these points are fullfilled, will the associate editor always start to look for appropriate reviewers?

Once I heard in a talk of a physical review letters employee they get 10.000 submissions per year and 1.000 articles are finally published. I cannot believe they start 10.000 review processes, this would be a waste of everybody's time. On the other side every physicist knows PRL is the most prestigious journal in the field and you cannot submit garbage. So I believe most of the 10.000 submissions fulfill the formal standards. Based on which criterions the associate editor will then not start the review process? Age/quality of references, abstract, his historic and topical background and expertise? I never was associate editor myself but I guess they don't read the full article, but skim abstract, introduction, figures, references and conclusion maybe?

Answer

I'll give you my perspective as an co-Editor-in-Chief and previously an Associate Editor of the ACM Transactions on Mathematical Software (ACM TOMS):

• 
  

  As Editor-in-Chief, I take a brief look at each paper and decide whether it even makes sense to move forward with it. I would say that between 10 and 20% of the papers are already filtered out at this stage: They are not within scope of the journal, or are obviously not at the level we expect -- we get our fair share of submissions from second and third world countries (and some from first world countries) that are at the level of semester projects for advanced graduates.

  
  


• 
  

  As Editor-in-Chief, many submissions I get are not within my research area. The ones already filtered out above are obviously not acceptable, but then there are many that I suspect are not good enough, but where I don't know the area well enough to make the call. These get handed on to the Associate Editors, often with a note that they should take a look whether I'm right or wrong in suspecting that the paper does not meet our criteria.

  
  
  


• 
  

  As Associate Editor (before I became EiC), I then reject another ~15% or so based on my subject area knowledge.

  
  

So I suspect that for this journal -- a fairly reputable one with one of the highest impact factors as far as math journals go --, about 1/4 of the papers get rejected without being sent out to review. Of course, then another 1/3 to 1/2 or so of the rest get rejected after review.