DATA ANALYST

In my experience, the job title of data analyst is vague so look at the tools being used to find out what the job really does (the further down this list then the more complex). In general, you'll find yourself getting, storing, cleaning, and analyzing data to make some sort of data-driven decision. Data is pretty powerful; remember that big companies like Google and Facebook are in the business of offering 'free' services so they can collect data on you and tell you what to buy. Here's how to get started.

Follow these steps

  1. Let's first learn Excel. Yes, it's boring, but most places use this and despite some haters, it's a really good all around tool. The key is that you *quickly* get to see that data is stored and you can apply functions to the data to get some type of output (more data, usually in the form of charts). A good introduction is 'You Suck at Excel' by Joel Spolsky.
    (https://www.youtube.com/watch?v=0nbkaYsR94c)
  2. You'll probably start off a data analyst job running some existing reports in your organization that shows some important metrics (e.g. number of ad clicks, satisfactory survey). You'll be able to make some changes to the reports, usually modifying or creating new data visualizations. This might be in Excel or some type of data software like Tableau. If you're super fancy, you can try something like d3.js. If you're not sure what chart to use for your data, look at this infographic to help you decide.
    (http://extremepresentation.typepad.com/blog/2006/09/choosing_a_good.html)
  3. Next you should learn a bit about relational databases and how data gets stored. Learning SQL and being able to use a tool like SQL Server Management Studio or pgAdmin (depending on the type of database you're using) is going to be really helpful in tying together the data you have. You'll learn about tables, selecting data, joining data, and how to input data into a database and how to extract that data out. Any of the database systems are great, just avoid Microsoft Access; you can thank me later. SQL systems are pretty much the same. Most of the time you won't need more than a few JOINs.
  4. So where are your skills at now? You can do a basic job of getting, inputting, and visualizing data. This is a pretty solid start for a data analyst position. You'll need to learn some domain knowledge (e.g. if you're at a hospital, you might need to learn a little about medical coding), but your base skills using tech to work with data are transferrable across a lot of domains.
  5. It's time to take a big leap and learn basic statistics. It's inevitable, but also it's NOT as hard as you think it is. I recommend the book 'Discovering Statistics Using R' by Andy Field. You will learn statistics and a statistical programming language. If you use a different programming language than R at work, I still recommend the book, but do the book exercises in your language (e.g. SPSS, SAS, Python, Julia). You will learn the types of data and how to model the data.
    (https://www.amazon.com/Discovering-Statistics-Using-Andy-Field/dp/1446200469)
  6. At this point you'll have a good understanding of what statistics is doing, but only in theory. Go out and do some practice with real life data, then check if you're using the right statistical model using a cheat sheet. Try to really understand WHY a test type is being chosen and if you can pick out the patterns across different data models. The only real way to learn is to use real data on non-tutorial examples.
    (http://www.ats.ucla.edu/stat/mult_pkg/whatstat/default.htm)
  7. Now that you know basic statistics, you can create your own data models. You know what data to capture and how to interpret results, yet you also know data's limitations (i.e. data will tell you anything if you torture it enough). You'll start to find that data analyst jobs aren't really technically challenging anymore. If you're interested in what else you can do with data, go look up the guide for Data Scientist to do some machine learning; this will lean towards more math and programming.