Are you sure that traditional sources of knowledge, and traditional ways of making people believe things, result in greater accuracy? Consider that over half of the papers in many fields, including medicine, are statistically bogus: http://journals.plos.org/plosmedicine/article?id=10.1371/jou... http://biorxiv.org/content/early/2016/08/25/071530

On average the results in science papers are probably more accurate than random stuff you read on the internet. But I suspect the difference in error rate is uncomfortably small, much smaller than those who are smug about science would like to believe.