It's well-known that most people in countries where the issue has been studied consider themselves to be above average, despite that only half of people can be above average, so I'm asking for a cutoff of 70% instead of 50%. I can't stop people from participating in the survey, but know that if too many people who don't meet this requirement participate, the data will be distorted.
As the survey notes, the smart people we know about come from various filters. If a famous scientist with one million followers seems to care about stupid people, it doesn't mean all scientists do. Keep that in mind.
I think the results for this question would vary significantly depending on country, but nothing I can do about that. Similar to trust: in China or Sweden,
60% of people say that most people can be trusted, while in Brazil or Columbia less than 10% say that. (Pew Research in 2007 found
79% for China, 35% for Brazil.)
I think in an ideal society, people would be able to say that yes, smart people do try to make the world better for stupid people. Whatever you respond on this survey now, if the world changes, you might change your answer to this question. In that case, this survey would allow a "before and after" comparison.