Let's do a study!
Actually, I think that the folks at the CDC and university medical centers all over the country (and in other countries) are acutely (even painfully) aware of the number of people who aren't vaccinated.
What GR and other advocacy groups have been agitating for is a study comparing groups that don't vaccinate their children (e.g. the Amish, although they actually do vaccinate their children) with vaccinated children. The folks at GR are pushing for this despite the fact that their baloney "survey" showed no connection between autism and vaccination.
Here are a few reasons why that study wouldn't work, followed by my own "modest proposal" for a study that would work.
 The unvaccinated population is not necessarily the same as the vaccinated population: People who don't vaccinate their children may not be the same as those who do. In fact, studies have shown that those who don't vaccinate their children tend to be at either end of the socioeconomic and educational spectrum. It would not be valid to draw any conclusions about the impact of vaccines on autism (or ADD/ADHD) with such marked differences between the groups (apples and oranges).
 Selection biases: Recruiting people for a study is one of the biggest sources of error, especially when looking for something that is a public controversy. You will tend to attract more people who believe in a connection than those who do not.
 Inadequate power: If the GR survey is correct, about 3% of the overall population in the 4 - 17 year age range is completely unvaccinated. On the other hand, the prevalence of autism is only 0.65% in the same age range. It will be easier to detect a statistically significant difference in unvaccinated children between autistic and non-autistic groups than it will be to detect a difference in autism prevalence between vaccinated and unvaccinated children.
 Loss of useful information: It would be a shame to spend the time and money to do a study and only be able to answer a single question. After all, if there is no connection between autism and vaccination - as the GR "survey" suggests - wouldn't it be nice to be able to "mine" the data for other possible connections? By looking for subjects based on their vaccination status, the only question that can be answered is whether there is a correlation between the two.
My Modest Proposal:
This is a study that could be done rather quickly and with a minimum of expense. It would eliminate many of the sources of bias and would fairly easily generate balanced study populations that would be a good match to most of the general population.
[a] Contact a large HMO with actual facilities (e.g. Humana or Kaiser) and arrange to get access to their patient medical records. This is routinely done, although the HMO will want assurances that patient confidentiality will be maintained.
[b] Obtain a list of patients with autism diagnosis in the proper age range (I would suggest 6 - 12 years).
[c] Select one thousand of these patients at random. This would allow you to detect a difference if the prevalence of unvaccinated children is less than 1/3 that in the general population (alpha error level 5%, beta error level 5%). If the difference is less than that, you'll need to select more subjects.
[d] Confirm the diagnosis by having a child psychiatrist or psychologist review the records.
[e] For each of the remaining children, select a non-autistic control child from the HMO database that is of the same age, sex, geographical region, etc.
[f] Determine how many of the children in each group have received all, none or some of their vaccinations (keep track of which vaccines, when, etc.). If the GR "survey" was right (a very big "if") about the number of children unvaccinated, each group should have around 30 unvaccinated children, unless there is a correlation between vaccination and autism.
[g] If the autism and non-autism groups have statistically significant differences in their vaccination rates, then a correlation can be claimed. If the study shows no correlation, then the relative risk is less than 3. You'd have to have twice as many subjects to bring the minimal relative risk to below 2.
Using HMO patients eliminates any issues of affordability (which is minimal) or access to health care. Although the population of people who have HMO coverage is not necessarily the same as the overal US population, using the case control design ensures that the two groups are as similar as possible.
This study could have been done in the time - and for the reported cost - of the GR "survey". If they had done a real study - like the one I've outlined above - the answer would already be here.
If anybody would like to fund such a study, please contact me. If I can't find anyone else to do it, I'd be willing to coordinate it myself.
I doubt that any of the federal funding agencies will be interested in funding this study - not because of any conspiracy, but because the only people who want the study will probably not believe the likely outcome (i.e. no association - just as in the GR "survey").
So, this is my challenge to those who want this study: if you want it, do it. The advocacy groups have the money - all they need to do is find the will to take a chance. And make no mistake - this is a risk. There is a better than even chance that the study - if it isn't "fudged" - will show no correlation.
If they really want to know, they can find out for less than the cost of a full-page ad.