Discuss how political scientists attempt to quantify data, manipulate data statistically, and attempt to validate hypotheses

There are a few ways that this can be done. I will list the ones I know off the top of my head, as this is a broad question. I will be describing this as if the reader is a political scientist so don't take it personally, it's just so you can better understand.

Cherry Picking: Say you want to get a poll. Don't go to an area where there is a majority of the working class. You wan't to go to areas that are high on welfare, stock brokers, corporate etc. You can best find these in big cities like New York, Orlando, so on. That way, when people see the polls, they will look at what are demographic is, giving them the sense that we are popular party.

Fabrication: Don't worry about the legitimacy. Most people just want some sort of facts or explanation, and if we make one up that it is believable for most people, we are good. So long as they don't look at the actual ice caps and the temperature around the world, as far as they know Global Warming is a thing, *cough* I mean, sorry not Global Warming, Climate Change, it's totally different *wink, wink*.

Censorship: So long as no one sees any contradicting evidence to our polls, they are good. Same thing with Climate Change. If however something does come up we need to treat is as if it is absurd. Conspiracy theorist!  But even then we need to make sure that it stays away from public eye, and public opinion. 

Hope this helps, don't want to keep you waiting much longer.