Andrew Weiss, Indiana, 60,000 Calls discussed on Indiana Issues


Little more realistic about what that means of Andrew couple mechanical questions for let you go here because obviously, whenever we do polling people say, Well, I don't trust polling or nobody called me or how can 400 people basically determine what 900,000 people think of walking through the science of what you guys did or how you did it? Yeah. So, uh, what? I don't want to get too technical because that would just bore people to death. But essentially, you know, from a statistical standpoint, what you're allowed to do is take a representative sample of a larger audience. And then from there, you're able to make inferences about that larger population. And the key is that it's a random sample. And so you're right. We didn't contact everybody. Um, Frankly, that would be sort of, uh, on, you know, first of all, from A cost standpoint, it would be expensive, but from a practical standpoint, obviously not everyone's going to answer their telephone. Not everyone is going to, uh, you know, answer a survey whether it's delivered in this instance from online or text message surveys, which is how we perform this one. And so again, we're able to get a representative sample. And so what we looked at was the overall registered voter population in Marion County. And we said Okay, for example, you know, we need to make sure that our Gen Er is about right. And so it needs to look like the overall population and so registered voters in Marion County. It's about 53% female about 47% male, and so we look to balance that out and ensure that our sample again this random sample is representative of that we need to ensure we have enough young voters enough older voters enough white enough black voters and then also from sort of a geographic area as well. And so when you look at that Statistically speaking, This is a sound practice and, um, measure of what the larger population looks like. But you know, there is a margin of error. And so that's where that 4.9% margin of error comes into play. Because statistics while great isn't perfect, But it is from a mathematical scientific perspective. Very sound waves measuring overall opinion. Something else you folks did, too. Which I thought was interesting was he did, uh The old posters where you Go pick up the phone and call call random folks are based on voter to voter data list. He did online and also text messaging. Uh, white Online. Why tax and why not just No go to go to the old random. You grab an old phone book. Remember, the phone book is and just start calling people. Yeah, well, so you know, one of the easiest ways to explain. That is cost. So, um Phone that you know, to get a live agent in Indiana. Or you do need a live operator to conduct a survey. You know, in order to make, uh you know, uh, yes. 400 Sample survey, you know, say 60,000 phone calls. Obviously, you know, you gotta pay the live operator agents to make 60,000 calls And so those costs add up, and, uh, That doesn't necessarily though, mean that it is a better product from a cheaper option, which again is serving text messages as well as online surveys. And so the reason we chose this was from a practical standpoint, it's just more cost efficient. And if you think about younger voters, actually minority voters, they actually they're easier to reach now online and via text messages anyways, you know, in 19 year old Kid probably doesn't have a landline phone. First of all, or pick up a phone call from some random number, but you know, they're sitting on a train there, sitting waiting at a friend's house who knows doing whatever they see a text message come in. They're very inclined to sit there and actually answer a survey. And so we find that the quality remains the same. But we're able to keep costs a little more efficient and Thought that pink. Um uh, recommendations coming out of sort of 2020 and pulling postmortems was a diversification of methods in how you conduct polling and the old gold standard of the live operator survey being conducted over the telephone still has its purposes. But again, it's a little more expensive. But you also you miss out on some of these voters who again don't have landlines or don't pick up their cell phones. Um you know young kids nowadays like Don't Even talk on the phone. It's all text message. It's all email that sort of thing. And so, um In order to ensure that random sample we look at various ways that we can reach out to people that are sort of outside the box and text message. Online. Surveys have been around for a couple election cycles now, but they are extremely accurate and viable way to conduct surveys that sort of reflect the changing times and societies practices and that's how it's going to ask you about the 2020 next 2016 elections as well. Because the polls are sort of kind of wrong and all over the place. How did you folks compensate for all the things we learned in the past couple election cycles? Well, so the good news is, you know, in an off year, you know people are left are more likely to answer a survey. It's really it's when you come down to the last two months. And, you know, people have had 15 surveys come through whether they're less likely to answer a survey. But, um, you know, I think what I saw in 2016 and 2020 is that Public polling and I'll talk about it in terms of public private, but the public polling that is the stuff that you saw largely, um, you know from Gallup, or, uh, you know, some of the universities are like, you know, NBC or whatever doing these national polls. You know, The problem with those national polls is that there's a lot of what I'll call sort of armchair polling critics and so they look at the methodologies and they'll go well. This is wrong, and so people are left. Apt to change and adjust based on what sort of the data is telling them and in 2016 1 of the takeaways was that You know, uneducated or I should say, uneducated, non college educated Excuse me, white voters and sort of rural areas where being up under represented, and so people wanted to see the education breaks and started to think. Well, it needs to be sort of X percentage of college educated voters. X number of non college educated voters. And people weren't able to deviate from those because these sort of pulling critics without well, that's that's not. That's not what it's going to be. That's wrong. And so people aren't able to change. I think as well as sort of the private pollsters, which look at the gate and what the data is telling them and say, Oh, well, actually, we see You know this trend happening or that trend happening? And so I think a lot of the private polling was actually pretty spot on. Um, it was more so some of the public polling where he saw some of the issues, but, um, you know, we we saw plenty of other colleagues pulling my own polling where we saw some of the data coming back. And you see things like, Oh, you know, A lot of these voters who we would never have expected as a quote unquote likely voter. Maybe they hadn't voted in the last four or five elections, all of a sudden expressing a strong interest in this election. It's Oh, well, Maybe we're going to see an increase in turnout. And so we're able to make some of those adjustments. Um, that may be a public facing, you know, sort of publicly scrutinized pole aren't necessarily able to make All right, Well, we covered a lot of ground today. Our guests the program's been Andrew Weiss, Sir Andrew is the principle of a RW strategy supporting from that We hired any politics to talk about the city of Indianapolis and to some of the big issues going in there..

Coming up next