[MUSIC] Welcome back. In this segment, we are going to talk about the impact of our findings on our initial hypothesis or hypotheses. And how what your metrics may show may not be exactly what is really happening. So let's kind of refresh ourselves, we had our project proposal that we talked about before, right? We started off with our questions to answer, what was our approach etc. And then what our initial hypotheses were going to be. So in my case, my hypothesis was that his tweets were going to be highly connected to Tesla and SpaceX. And that they'd be discussing energy and space related topics and that mostly a positive tone. We'll cover two and three in a different segment. And for the first one, we'll talk about that one today. So let's jump into our notebook. So if you recall, we were going through and taking a look at different metrics. We had public count, mention count, retweet counts, etc. So we had up here a whole list of public counts, but even with those public counts, the highest public count that also had a mention count was boring company and had a pretty high retweet count as well. So we'll take a look at that one. In mentions, we have Tesla and SpaceX which is what we expected to see, but we also see this Everyday Astronaut kind of in the middle here. And not a lot on the retweet, so we'll kind of take a look and see what is this one really about. And then taking a look at retweets, we have Tesla and SpaceX at the top which we expected. But then we also see Tesla Rowdy, so we'll do some investigation of that as well. So we're going to jump down and take a look at what we've done for analysis. But before we do that, let's get a quick grounding on our ERD again. Recall we have we Tweet, which is our main entity which houses all of our tweets and then we have broken out tweet user which has the tweet ID associated with a specific username. Which is what we're going to be interested in right now as we want to take a look at a few different tweet user IDs. So what we've done here is we've got a few data frames that are specifically created just to look at Everyday Astronaut for one just SpaceX tweets and one for Tesla Rowdy. We're doing a little manipulation here where we've gone through and modified the created ads so we can take a look at those. So here we've got some some plots, some distributions of tweets. This one is for the Everyday Astronaut tweets those that were mentioned where Elon Musk mentioned Everyday Astronaut. And it looks like this is somewhat isolated, right? So the time frame was from April 2018 to towards the end of 2019 here. But if we do a little bit of research we also found that in October of 2018 Everyday Astronaut did an interview with Elon Musk, right? So that kind of accounts I think for this engagement. Well, how does that compare if we look at what his tweet pattern look like for SpaceX. So we see a very different distribution and it's not within kind of the month time frame, as we're looking more at the yearly time frames and it's consistent all the way across time. If we look at Tesla Rowdy, we see the same kind of thing. So Tesla Rowdy I think is legit. That's one, that's kind of new and pops into the middle of our Tesla and SpaceX. But Everyday Astronaut I would almost look at that as a one-off engagement that took place but not something that he's been engaged with for a long period of time. So what does that mean for our overall hypothesis? Well, just as we said our data shows that these were very high. However, we also saw high mentions for Everyday Astronaut in retweets of Tesla Rowdy in connection with boring company. So we've already I think somewhat eliminated the Everyday Astronaut from our overall analysis thinking that that's something that he is talking about over and over throughout time. But it does connect with his SpaceX type work. Tesla Rowdy though does exhibit more of a similar distribution that of SpaceX. So that one I think is up there and we would have to add that in to our hypothesis and probably do a little more investigation as to what Tesla Rowdy is all about. The high public tweets of boring company and the mention counts of boring company kind of makes sense because this is the hyperloop, I have a typo here. Hyperloop Loop company, that's the subsidiary of SpaceX and as of 2018 Elon Musk owned about 90% of the equity interest in that company. So it kind of makes sense, he'd kind of put those things out there as well. Okay, well, that's just one aspect that we're taking a look here of the impact of our findings. Something to just keep in mind as you're looking at your data the metrics may show you one thing but you might need to dive a little bit deeper to make sure that you're not seeing some kind of outlier or something that maybe can excuse what your data is. Sitting in front of a customer you may get that question that says, hey, what's the next level down? And you're going to want to have that answer as well. Okay, until next time.
