• 3 months ago
It's so easy to lie with data!

In fact, it happens all the time, but we just may not realize it (cough cough *all news networks*)

In this series I show you some of the most common ways to lie with data!

____________________________________________

RESOURCES:

đź’»Analyst Builder - https://www.analystbuilder.com/

đź“–Take my Full MySQL Course Here: https://bit.ly/3tqOipr
đź“–Take my Full Python Course Here: https://bit.ly/48O581R
đź“–Practice Technical Interview Questions: https://bit.ly/46pDqqL

Coursera Courses:
Google Data Analyst Certification: https://coursera.pxf.io/5bBd62
Data Analysis with Python - https://coursera.pxf.io/BXY3Wy
IBM Data Analysis Specialization - https://coursera.pxf.io/AoYOdR
Tableau Data Visualization - https://coursera.pxf.io/MXYqaN

Udemy Courses:
Python for Data Science - https://bit.ly/3Z4A5K6
Statistics for Data Science - https://bit.ly/37jqDbq
SQL for Data Analysts (SSMS) - https://bit.ly/3fkqEij
Tableau A-Z - http://bit.ly/385lYvN

*Please note I may earn a small commission for any purchase through these links - Thanks for supporting the channel!*
____________________________________________

BECOME A MEMBER -

Want to support the channel? Consider
Transcript
00:00What's going on everybody? Welcome back to another video. In this series, I show you how easy it is to
00:04manipulate and lie with data.
00:12I don't know about you, but if you've watched this entire series on how to lie and manipulate with data,
00:16it really makes you trust almost nothing, and I think that's a good thing.
00:20I don't think we should trust anybody or anything that is online, but that's just a side note.
00:24In this video, I'm going to show you how to create biased surveys to make your data look really good.
00:30All right, so let's take a look at some survey data. Now, I am always very wary of survey data,
00:36and I'll tell you why.
00:37It's because surveys are so easy to manipulate, it's not even funny.
00:42And I'll give you an example back before I was even a data analyst. This was back when I was
00:46in college.
00:47I was in a rec therapy program, and we had to conduct surveys for one of our classes.
00:51And so what we were initially conducting a survey on was if people liked going to the gym.
00:55And naturally, we went to the gym to ask people. And when we got our results back, our numbers were
01:01extremely high.
01:02People love going to the gym. I mean, it was like 95% of people liked going to the gym.
01:07And we're like, whoa, people really like going to the gym.
01:10Until we started thinking about it, we're like, wait a second.
01:13This probably isn't very representative of the population on campus as a whole, because not everybody actually goes to the
01:19gym.
01:20Maybe we should go to a more neutral place, like maybe the dining hall, where we would just get a
01:24myriad of people, just random people, and then we could conduct that data.
01:28And then we found that it was only like 30% of people liked going to the gym.
01:32Now, really quickly, let's take a look at this survey that was conducted.
01:35We have a social media survey number, and we have a like, dislike, and they were just neutral.
01:41Now, you can just immediately tell by this survey that people really liked whatever this survey was about.
01:49They loved it.
01:50Out of the 1,000 people who responded, 900 of them liked it.
01:53Only 50 people disliked or were neutral about that post.
01:57This is something that they're going to post everywhere.
01:59This company is going to be like, look how much people absolutely love whatever we were talking about.
02:03But, as you can imagine, they posted this on their social media, where people who already like them are going
02:10to follow them.
02:11So, now they kind of have this echo chamber of, we really like what this is and what you're doing.
02:17And so, of course, naturally, you're going to have a lot less dislikes, a lot less neutral.
02:21But let's go over here to the bias survey good.
02:25This is where we do kind of a more representative or larger scale survey.
02:29And this is going to be for all users.
02:32This isn't just people who follow their social media.
02:34This is everybody who uses their device, their platform, or their app, or whatever it is.
02:39Now, if you ask a larger population, people who don't follow them on social media, there's still a good amount
02:44of likes.
02:44There's still 2,000 likes for this poll.
02:47But you'll notice that the dislikes and the neutrals were a lot higher.
02:50We have 1,000 for each.
02:52Now, this is still good for this survey.
02:54This is still a good thing.
02:55But this is most likely more representative as a whole than just asking their social media.
03:01But you could very easily take this and run with it and say, people love this.
03:05People are overjoyed with this change that we made or this new product or whatever it is.
03:10But that may only be true for your core users.
03:12This may not actually be a good thing for maybe your outside users, for people who are interested in the
03:17product, who might be looking to buy your product.
03:19And so there are different surveys that you need to conduct for different uses.
03:24Maybe there's a change that you're making for current customers who follow you on social media.
03:27This is going to be a great survey, very representative of your exact population that you're trying to get polls
03:32from.
03:33If that's what you're trying to do, and then you ask everybody, then these responses are going to be skewed
03:38as well.
03:39So knowing your target audience, knowing who you should be asking, who you shouldn't be asking, or including in these
03:44surveys is very important.
03:46Again, I've made this mistake myself, and in fact, most of the mistakes that we've talked about in this series,
03:51I've made myself.
03:52And so I'm not above these things.
03:54I'm just learning from experience and sharing it with you.
03:57So I hope that you learned from this.
03:58Don't be that guy or girl who does this and then messes things up and you look like a fool
04:02in that next presentation.
04:04I want you to look good.
04:05So if you like this video, be sure to like and subscribe.
04:08And I will see you in the next video.
04:20I'll see you in the next video.

Recommended