Publishing preprints on sites like SocArXiv can be negatively viewed by non-open traditional journals (treated as prior publication - this has happened to me) Might this also be a relevant factor? Ie disadvantaged groups prefer conservative journals because they feel less able to take risks?