When I have done workshops for aspiring science journalists/writers, I have three pieces of advice. The first is: Don't defend science. It doesn't need defending.

But it's easier said then done. If you spend some time in science media culture, you will invariably find a person saying something pithy like "Science: It works, bitches" but then raging about some attack on science and defending it with shrill verbage and name-calling and conspiracy theories.

If science works, you don't need to defend it with claims that Big Oil is funding climate denial or that homeopaths and Big Organic fund vaccine and GMO denial.

Biologists and farmers do not want labels put on GMOs because a special label lends a perception that there is something to worry about. Labels are why "No BPA" and "Gluten-Free" products are all the rage, not evidence. BPA has harmed no one and gluten-free foods charge 242% due to the perception of better health, when only 5 percent of the people buying gluten-free foods are gaining a benefit. The others are likely harmed.

This ad doesn't hurt, according to a new study. California leads the nation in anti-science beliefs, but the anti-vaccine craze is the really dangerous one. However, the same ad debunking any link to autism might do more harm than good.

Lyndon Johnson had one of the more famous examples of using this labeling tactic. In 1948 he was trailing the incumbent in his first Congressional election with just over a week to go. He instructed his campaign manager to spread the rumor that his opponent's family wanted him to stop having sex with their farm animals. His campaign manager said it was crazy, no one would ever believe it. Johnson replied, of course not, I just want to make him deny it. He closed a 10-point deficit and won.

Making companies publicly defend problems they never had - "have you stopped beating your wife?" - is a time-honored tactic.

So it seems like it would be a bad idea to proactively declare you have stopped beating your wife. Yet that is what vaccination PSAs debunking a link to autism might be doing. 

Weirdly, this book by osteopath (what a surprise - just like homeopath and anti-vaccine crank Joe Mercola) Sherri Tenpenny lists herd immunity as a reason not to immunize your child. Yes, she appeals to people to protect their special snowflakes by letting poor parents be good little government drones and immunize their kids to protect the elites who buy this book. If you want to know why the coast of California is made up of so many anti-science crackpots, this is a good datapoint.

Messages showing that no evidence for a link between vaccines and autism may be lending credence to the belief that there is, in a "where there is smoke, there is fire" way. During June and July of 2011, 1759 parents with minors in the home got one of four messages. Three dealt with diseases that vaccines prevent and the benefits while one discussed the lack of evidence that the MMR vaccine causes autism.

None of the ads caused parents to be more likely to vaccinate a child. But the ad refuting claims of a link between vaccines and autism caused parents less inclined to vaccinate to believe in vaccinations even less. Attempting to correct misperceptions about crackpot vaccine claims were counter-productive. Lyndon Johnson could have told them that.

This ad may not help with vaccine uptake - 'I don't know a kid with polio' is an obvious reason sane people already accept vaccines - but at least it doesn't hurt. Dignifying a debunked study with n=12 by doing a PSA against it seems to hurt acceptance.

There is no question we don't want to ignore the political and cultural cluster that has adopted the anti-vaccine mentality. In Science Left Behind (now out in paperback!) I somewhat joked that you could simply look at the parking lot of a Whole Foods store and make an accurate inference about science beliefs and voting records and lots of other cultural things. Now the correlation is so strong it isn't remotely funny. Measles are back, Whooping Cough is back and at least one school on the California coast only has a 26% vaccination rate - far below the ~95% rate needed to maintain herd immunity that the more scientifically literate populations of states like Alabama and Tennessee know and embrace.

It's just smarter to remain positive - and not be heavy-handed. So don't hire someone who wants to roll out a "Vaccines: They work, bitches" campaign.

Citation: Brendan Nyhan, Jason Reifler, Sean Richey, Gary L. Freed, 'Effective Messages in Vaccine Promotion: A Randomized Trial', Pediatrics March 3 2014 doi: 10.1542/peds.2013-2365