Determining true results with contradicting results. (Science)

[IceSage]

So, I've run into a kind of rough patch with the whole "vaccines and autism" debate. I'm unable to properly explain to someone what sources are credible and what aren't. Also, I'm unable to clarify how one actual scientific source can be more credible than another.

A few of my friends are 100% Conspiracy bogosity drones. They read on the internet that "Vaccines cause autism" in an actual scientific source, and use it to say, "Here, it does."

If I point them to another source that debunks this, their claim is "What makes that dismiss the scientific claim that says it does?"

I need some help in constructing a critical thinking set and guidance on how to explain why I believe the common scientific and accepted documents and papers in science, out-weight the links they hand to me after performing some Google-Fu.

At this point, it's getting confusing to me myself. I stopped by to see if Shane or anyone could explain it better.

[Ibrahim90]

I'm assuming the paper they use is the one from the lancet? if so, you have to demonstrate how the researcher arrived at his results (hint: he wasn't the most ethical), and how the journal retracted the article, the author lost his license, etc.

put simply, the man is guilty of fraud (and murder).

also, it is one paper only (any other source I'm aware of is simply not from a scientific journal/a science article): there are dozens of papers on the other hand that say otherwise.

additionally, point out that FAR more people are vaccinated, than get autism (autism only accounts for ~1% of people; compare that to the numbers vaccinated). if there were a one to one proportion, it would be much higher (and if they say the 1% (or a part of it--doesn't matter) were from vaccines, point out how autism has become more common statistically in spite of a drop in vaccinations).

finally, point out to them that the real risks of death from preventable illnesses are invariably more severe in their toll than vaccinations: at least an autistic child can be treated: who can treat the dead?

and if they be religious (I'm mostly dealing from an Islamic perspective), remind them that their religion demands that medicine be not forsaken. Jesus healed lepers and the blind: this is a pittance compared to just keeping hundreds of thousands healthy with a vaccine.

if they won't listen to reason, then they are stupid. and murderers. and you have no need of their "friendship".

(speaking of which, need to unfriend someone on facebook over this. I will not tolerate this "belief" one bit)

[IceSage]

He links me stuff like this bogus website that actually link to seemly credible sources.

I'm just having a hard time trying to tell him that some of these sources aren't credible or are flawed results.

http://www.activistpost.com/2013/09/22-medical-studies-that-show-vaccines.html

Like this: http://www.ncbi.nlm.nih.gov/pubmed/21058170

Then claims, "Oh look, it shows this can cause autism. Vaccines cause autism."

[Ibrahim90]

can't read it, so cannot judge the paper itself.

having said that, from what I do see a couple of issues:

1-only boys are tested
2-the paper seems to be somewhat similar to a 2008 paper regarding demyelination disease (which isn't autism, nor does it cause it, as I recall). This paper is legitimate, but here's the thing: there are two different Hepatitis B vaccines: only one (the less common one) is an issue. and its being (or has been, depending on the country) phased out.
3-the ages are generally of those who should have already been vaccinated (3-7), not when they start vaccination (you do this to kids under 2 years old: this is especially if the mother is known to have it prior to the child's birth): by this point, symptoms of autism will already start showing: this is the biggest problem, since it could be the case that there was a sampling error (and I'm being generous).

[dallen68]

Firstly, I'd look in as many publications as I could to see what the general consensus on the particular issue is. If there are many hundreds of articles saying the vaccination (in this particular case) and a few dozen saying it isn't - Then it can be reasonably surmised that it probably is. Even discounting deliberate dishonesty, as alluded to by Ibrahim, anomalous results are not uncommon due to a variety of factors, such as data entry error, research bias, unknown pre-existing condition, and so forth.

Secondly, if there's a 1:1 chance of the hepatitis vaccine causing autism, that would mean that every single person that gets the vaccine would get autism (just so you know). We know that doesn't happen, with the 1:100 ratio cited by Ibrahim, the vaccine wouldn't even be looked at by any body who wasn't pursuing an agenda as a probable causation. Generally speaking, medical researchers look for a 1:25 or 1:30 ratio before they start claiming correlation (which isn't the same as causation), and about 1:15-1:10 before claiming "possible causation", and then in depth research begins at that point.

Thirdly, look at what the NCBI report ACTUALLY says. All it says, is in an isolated study boys who received the vaccine at a certain point were "three were three fold more likely to develop autism" than those that never got it, or those who got it at the regularly scheduled time. Fear mongers exploit language like this all the time, and it doesn't help that the portion of it there doesn't tell how many boys were in the study. For sake of argument, let's say 10,000. So normally, say, you have a 1:10000 chance of autism (regardless of anything else), in this study 3:10000 developed autism, which would be well within the range of a normal variance (it could just be a co-incidence). However, you'd actually have to find the actual paper the extract was from to determine the exact context of the statement.

*I am not claiming there's a 1 in 10000 chance of autism, that was only to illustrate how the "three fold" thing isn't as scary as it sounds there.