Exactly so. Sample sizes of 300 - 500 respondents are sufficiently large to be able to project the results of the survey to the broader population…if the sample population has been randomly selected, and is reasonably representative (proportions of age, sex, income, region, etc.) to the broader population.
Another problem is that even if you make a perfect random selection, the people you select may not all want to participate. I do not know how they design this type of survey. But I know what you *can’t *do is just post a poll and let people respond.
“Non-response error” is a big (and increasing) issue in market research, but probably takes the topic into an even more serious sidetrack from the OP. 
That said, “just post a poll and let people respond” is the sort of thing that we call a “convenience sample,” and results of those sorts of polls are generally worthless.
The one thing that polls will consistently be inaccurate on is the opinions of those who refuse to take opinion polls. But if you assume that such people are randomly distributed with respect to every other opinion, it doesn’t particularly matter. However, that’s a pretty bad assumption, so what you’ll try to do instead is have a survey that asks for all sorts of demographic information to try to make sure that the samples being chosen are as representative to the population as possible given the fact that various demographics are going to respond at different rates. Now all you have to do is figure out what the correct demographics are for the population as a whole. Good luck.
All advertising is false.
Yep. I mentioned this thread to my gf, who works in advertising (agency VP, project management). She rolled her eyes and said, “it’s all false”.
Your quote is missing the second line:
“to being eaten alive by fire ants”
Must be a very , very bad restaurant, if 1 in 5 will choose the ants.
Even though it is clearly a joke, I am wondering why the restaurant thought that basically saying “Our food is so tough it’ll break your teeth” was a good joke.
They didn’t say why the dentists supposedly preferred them.
I think they should put a Novocaine Special on the menu – cream of tomato soup, scrambled eggs and grits. Maybe throw in some jello for dessert.