Part of the problem with the media's representation of science is that they don't understand it themselves. One of the most most fundamental rules - "correlation does not equal causation" - is blatantly disregarded in media coverage. I find this especially true for longitudinal or cohort studies, where they are trying to identify risk factors, but then the media reports every other day that drinking a glass of wine will extend your lifespan/shorten your lifespan.
Science could do a better job explaining things and translating things into a way that people can understand and people need to become better consumers of science and educate themselves on how research is presented. It is a two way street to ensure that we start winning!
That's true, regular media is also willing to cover topics they do not typically handle and journalists writing about the topics rarely know very much about the topic they write about. At worst, they are just copying another article of the topic and dropping a part out.
But at best, the articles can contain references to the original study so readers who are interested can find the studies and see through them themselves.