Excerpts:
* 13,388 respondents
* of which 16% said Western Europe was their region of greatest familiarity
* of which 3,790 scholars were based in Europe
* of which 69% were academic staff, 15% research staff, 7% senior institutional leadership, 6% graduate/post-graduate students, and 2% teaching staff
* All in all, the average respondent reported spending 52% of the worktime doing research, 31% teaching and 18% administering. While senior institutional leadership spends 48% of the time administering, and 32% doing research, research staff spent 81% of the time doing research, teaching staff spent 57% of the time teaching, and graduate/post-graduate students spent 81% doing research, 13% teaching and 6% administering. Academic staff, the majority group (9,219 respondents), spent 46% researching, 37% teaching and 18% administering.
* of which 78% were male - most (87%) in Engineering & technology; fewest (69%) in Social Sciences
* of which only 779 identified as Arts & Humanities, the smallest group (Social Sciences and Life Sciences were both represented by 2000+ respondents)
* Research papers published: In average, a respondent said to have published 53.4 research papers (median: 30; meaning that a few publishes a lot and increases the average). Those with tenure had in average published 64.9 papers (median: 40), those without tenure in average 29.2 (median: 15). Even among those without tenure, however, 5% had published 100+ (with tenure: 17%). Males had published in average 58.9 papers (median: 32), females 33.5 (median: 20) - note that the survey says nothing about age distribution, which might explain part of the gender gap.
* Of senior institutional leadership, who had an average of 101.7 published papers (median: 70), 59% had published 50+ research papers and 34% 100+. Of academic staff, 33% had published 0-20 papers and 31% 20-50. Of research staff, 60% had published 0-20 papers, and of teaching staff 61%. Of graduate/post-graduate students, who had an average of 8.6 papers published (median: 6), 94% had published 0-20 papers and 6% 20-50.
Thomson Reuters remarks that respondents from Arts & Humanities were poorly represented in line with the poor publishing frequency compared to scholars from harder sciences (I wonder how the numbers would look like if a distinction was introduced between single authorship and co-authorship; or if the number of pages published had been counted...).