Why bother? Low effort data falsification is enough to land a high-ranking paper. Sure, really simulating data with some noise would be much more convincing but

Why bother? Low effort data falsification is enough to land a high-ranking paper. Sure, really simulating data with some noise would be much more convincing but its just not necessary. Raw data is very rarely inspected during peer review, much less so after publication, and most reviewers do not address statistics — if they do, the usually refer to their pet model and/or use the same incorrect tools as most psychologists.