Ah, science

One journal in medical research estimated that some 80% of published articles were later withdrawn, corrected, or contradicted by later articles.