…with this headline from TechCrunch: Study: Internet radio reaching 32% of households. Seems harmless, except the actual survey, from L.E.K., was an online survey that made no claims to national representation (Pew estimates that 26% of American adults aren’t even online) and correctly refers to its sample as “respondents” throughout the survey. At least 95% of the time, there is nothing wrong with the surveys that journalists cover, it’s the reporting that comes up short.