00:00What's going on everybody? Welcome back to another video. Today we're going to be learning all about
00:04statistics. Now using statistics is really relevant in just about any data job. I've worked
00:15as a data analyst for many years but I also worked with data science people and data engineering
00:19and there was a lot of statistics concepts that I've learned throughout the years. And so it's
00:23not just data analysts who use statistics, there's lots of other jobs and roles that use statistics
00:27as well. Now we are about to dive in into an entire series on statistics and that's going to be
00:32five
00:32or six videos. We're going to start with really basic concepts like mean, median, and mode but we'll
00:37work our way up to things like sampling data and even understanding confidence intervals. Now full
00:42disclosure I don't have a background in mathematics. Everything that we're about to learn I mostly just
00:47learned hands-on as a data analyst over the past seven or eight years. I actually have a degree in
00:51recreational therapy which again has absolutely nothing to do with mathematics but when I was in
00:57high school I took a lot of AP Calc classes and trigonometry and so I was good at math even
01:02before I became a data analyst. I loved math that's just not the career I had originally chosen but as
01:07you know I switched careers to become a data analyst and now I've been working as a data analyst and
01:11an
01:11analytics consultant for almost 10 years. So with that being said if you're looking for something that's
01:16like academic there are going to be lots of other statistics channels on here. This is for more
01:20hands-on practicality of using statistics within the data world. Now before we dive into this first
01:26lesson I want to let you know that I have an entire statistics course on analyst builder which
01:30you can go and check out in the description. It covers everything we're going to talk about in this
01:34series and it goes really in-depth and really hands-on and covers a few topics that we're not
01:39going to discuss here on YouTube. I'll have a discount code if you want to check that out in
01:43the description but without further ado let's jump into our first lesson. Now what we're going to talk
01:47about in this first lesson is just why we care about statistics, what it is, how we use it as
01:53data
01:53professionals and some of the core concepts that we're going to be taking a look at and that I
01:57really think that you should understand and know especially as a data professional. Now when we're
02:02working with data we work with a lot of raw data and raw data is just data as it sits
02:07within the
02:07system you're using it or however you're collecting it. It hasn't been cleaned it hasn't been transformed
02:12and you haven't applied statistical concepts to it to actually make use of that data. Now when I'm working
02:18with data and this is kind of just a side note but I still wanted to discuss this but when
02:22I'm
02:22working with real data I am almost never going to use the raw data. Typically this is how it looks.
02:28We start with our raw or staging database on the left hand side. This is just where we import the
02:34data from whatever source we're getting it from. It could be an API, it could be another database,
02:38it could be flat files or CSV files or whatever it is and we just bring it into our database
02:42but then
02:43we have to start cleaning it and we have to start you know digging into that data and once we
02:48do that
02:48we're putting it into our development database. Once we validate that maybe we have a QA analyst
02:53look at this then it'll move into our production database. This is where we start connecting it to
02:58a lot of our different systems and tools. So here we might start building dashboards off this production
03:03data. We might start bringing this data into different tools that we are going to look at in
03:07just a little bit. But this process is really important because typically you're not just going to be
03:11using that raw data. That data is just typically not super usable for your end statistics that you are going
03:17to be
03:17using. So now let's talk about how different data roles use statistics and this is just from my
03:23experience. You could use some other tool that I've just never heard of or some third-party tool
03:28or even certain domains have tools that are really common in that domain like maybe in banking
03:34that I just personally haven't used. So this is just kind of in general. Now the first one is probably
03:38the most obvious and that's just in spreadsheets and you can absolutely use statistical concepts and a
03:44lot of just generalized statistics within a spreadsheet just within Excel or Google Sheets or whatever it is and I've
03:50done that a ton. But what I will say is that Excel kind of limits some of the things that
03:55you're able to do
03:56within a spreadsheet. Now that has been somewhat updated with the ability to add Python within something like Excel
04:04because you can do that now. But again, I wouldn't say it's extremely common to use more advanced statistical
04:10concepts within spreadsheets. Now another tool that you're probably going to use is SQL. And SQL is how
04:16you're going to query your data. Now again, you can do this in a spreadsheet, but you do it in
04:20a different
04:21way and a SQL database is going to be able to hold an immense amount of data compared to a
04:26spreadsheet
04:26which has a limited amount. Now the two that I really want to focus on while we're here is Python
04:32and R.
04:32Now R is literally built by statisticians for data modeling for statistical modeling and for research and I have
04:41personally used it many years when I was on a data science team and there are so many things that
04:46you can do
04:47especially stats wise within R that you just can't do within something like a spreadsheet or SQL or BI tool.
04:53Now Python
04:54has a lot of similar abilities as R, although it wasn't specifically designed for it like R was.
05:02I myself prefer Python and I use Python. And in these next several lessons, once we start diving into it
05:09and
05:09getting more hands-on, we are going to be using Python because I could use a spreadsheet to kind of
05:16explain
05:16some of these things, but we're not going to be able to get into more of the advanced things. And
05:20that's what
05:21I want to dive into. And so Python and R are probably two of the best options if you want
05:26to really dive into
05:28statistics within the data world. You of course can use a BI tool like Tableau, Power BI, Looker, just
05:33common BI tools that a lot of companies use. These are more for kind of the end product of your
05:39statistics. You're typically going to have your data in SQL and you're going to be, you know, using it and
05:45getting it ready and cleaning it up. And then you'll read in that data with something like R or Python
05:49or
05:50wherever that data sits. You can read that data into, you know, these programming languages and do a lot
05:55more programming heavy statistics on that data. And then within those programming languages,
06:00if you want to visualize that, that's great. But to share your statistics with the world or with
06:06other people, your company, sometimes they don't understand Python and R. You have to simplify it a
06:11little bit. And that's where something like Tableau or Power BI would come in. You would simplify your
06:15findings and you'd put it in there so that you can share your dashboards and other people can then
06:20look at your analysis. Lastly, we have stats software. And this could be anything. I personally
06:25have used SPSS many, many years ago. But there's so many stats software or statistical software
06:31specifically made for stats that you can find out there. And this just depends on your company and
06:37what they may have. But these are a lot of the tools that you will use to use statistics and
06:43really
06:43dive into the data and then share that data with the world. Now, these are some of the core concepts
06:49or
06:49things that, you know, you should know ahead of time before we start this series. We'll learn how
06:54to calculate a lot of this by hand because I think understanding how to calculate it yourself instead
06:59of just using the function that's built in is actually really important because there's a lot
07:03of nuance depending on your data. Now, with sampling, there's lots of different ways you can sample
07:08data. And we'll take a look at different sampling methods. But we're going to start with the top left.
07:13We have mean, which is just the sum of all values divided by how many there are. This is what's
07:18called
07:18an average. Next, we have the median. And median and average or median and mean often get kind of
07:25confused or messed up. And this is a really important concept because median is the middle
07:30of your data. It's exactly cut in half. And oftentimes people think that's your average,
07:35but they are two totally different things. And once we get into the next lesson, we're going to get
07:39hands-on with mean, median, and mode. That'll be kind of our first hands-on lesson. I'll show you how
07:44these are calculated and why they shouldn't be considered the same thing, even though they get
07:49mixed up a lot. Lastly, we have mode. And that's the value that appears most often in your data
07:54sets. Those are what I would consider just the big three of simple statistics. This is like your basic
07:59level statistics. That's why it's going to be kind of our first lesson or hands-on because you have to
08:04know this to kind of go further. Next, we have something called a spread. And that's just how far apart
08:09your values are. Are they packed really tightly together on something like a bell curve? Or are
08:16they really spread out? And how can you determine that? What statistics can we use to determine that?
08:21And why do we even need to know that? We're going to dive into that in its own lesson. Next,
08:25we have
08:25standard deviation. And this is the average distance each value sits from the average. And standard
08:31deviation is super important and actually affects a lot of things down the line in statistics. Again, a really
08:37important core concept to know. And then sampling is when you look at a smaller group of your data set,
08:44you analyze it, and then you can draw conclusions based off of your entire population. Is it 100%
08:51accurate? Not likely. But if you don't have access to all of your data or an entire population of people,
08:58sampling becomes extremely important. So that is all we have for this introduction to statistics series.
09:05We're going to cover a ton in this series. I love getting super hands-on, making it really practical.
09:11In the next lesson, we're going to dive into that beginning part, which is the mean, median,
09:15and mode. This is something that once you really understand it, you'll be able to apply it and
09:19really just know what you're doing when you're working with your data for basically the rest of
09:24your career, which is a huge, huge thing. So thank you guys so much for watching. If you like this
09:29video, be sure to like, and subscribe, and I will see you in the next lesson.
