This is more than a replication error. Large-scale studies, with results that have too many variables, have strong tendency to underwrite the consensus. That

This is more than a replication error. Large-scale studies, with results that have too many variables, have strong tendency to underwrite the consensus. That wouldn't be such an issue, if they have huge impact on policy making. In this case the zeitgeist is against meat, for obvious reasons, but there are many more of them.