Many journals choose single anonymized review processing instead of double anonymized. I wonder what are the considerations behind these decisions to choose the single anonymized review?
The big advantage of single-blind review is ease. You do not have to anonymize the text before sending it for review (even anonymized, it's reasonably easy to identify the authors, especially with Google It's even trivially easy if the authors upload a preprint.).
The other main reason, I think, is less noble. At some level people want to base their decisions on secondary factors like the identity of the author/reviewer. Examples are:
- If the author is a well-known crank, then one might want to desk reject even if the paper at surface level looks like a serious piece of work.
- If the author is a Big Name, one might be more inclined to agree to review the paper.
- If the authors are undergraduates, one might be more lenient.
- If the author is an expert on the general topic, then even if the paper is radical, one might be more inclined to accept it.
Which is of course how bias manifests itself. But when things really boil down to a judgment call, it's extremely tempting to seek this information.
Your question is based on the wrong premise: That journals make an explicit decision to go with single-blind protocols. In reality, however, they use this approach because that's how it has always been done -- one would need to make an explicit decision to move away from this approach.
So the question then is: Do journals see sufficient benefit to switching to double-blind reviews to justify the added complexity in workflows for authors, editors, and support staff. Apparently, a substantial number of journal editors do not see sufficient value in this. (Whether that is an objective assessment of the situation may be debatable. But the decision is made by people, and so at the end of the day, the decision is made based on the subjective assessment of an editor or publisher.)