"I only took basic stats courses in college, but I know enough to know that's nowhere close to a big enough sample size to be statistically significant." I am

"I only took basic stats courses in college, but I know enough to know that's nowhere close to a big enough sample size to be statistically significant." I am not a statistician, but I'm pretty sure you're wrong about that. The real problem isn't the small sample size (2 million is really large), it's probably that they're not representative.