SQL and Data Science Fundamentals

You'll learn the basics of SQL and databases using PostgreSQL and you'll have the time of your life doing it. We use a real set of data: the analysis data from Cassini's flybys of Enceladus. There might actually be life up there and you're going to query the data to find out!

As a new DBA, You'll be given the raw data that Cassini gathered during its time orbiting Saturn and passing by Enceladus. You're going to load this data, transform it into a solid relational design using PostgreSQL, and then export it for analysis by the analytical team. When you're done, you'll be able to create tables and views, run analytical queries and tweak data as needed. You'll be on your way to your new life as a data pro.



You're about to dive into the glorious world of databases and the language of data - SQL. Lucky you! Our working data set will be the actual data from the Cassini mission, specifically that of Enceladus, an icy moon orbiting Saturn that is confounding scientists.

Installing PostgreSQL

We'll be using PostgreSQL for our SQL and analysis tasks. You can use other systems and they work just fine - but for this one we'll be using PostgreSQL. We have a bunch of data to load up and poke through, so let's get rolling quickly OK?


Creating Our Workspace

When working with data you often hear the terms "extraction, transformation and loading" or "ETL". This is something that data analysis people think about more than say Application Developers - which I think is a bummer. It's important to know how the data in any application is going to be used so you can make sure you collect the right stuff!

Importing the Master Plan

Our first task is to find the exact dates and times when Cassini flew by Enceladus and made its measurements. We need to create a time window so we can narrow down the results from the INMS - Cassini's on board "Ion Neutral Mass Spectrometer" - that's the thing that sniffed space for the chemicals we're looking for. Off we go!

Inspecting the Master Plan

Now that the data is in the database, let's poke around and see what we have, using simple select statements and getting to know the Postgres client tool psql.

Tangent: Working with Dates

Dates and timestamps are core to working with data as you will often find that if you don't know precisely WHEN something happened, it will become meaningless. Dates mark changes over time - those changes will often drive business decisions, so you better be correct!

Validating the Master Plan

The data in our mission plan looks straightforward but since it's a plan that is based on dates, we need to jump right into validating those dates. Thankfully for us, Postgres is outstanding at date and time functionality!

Creating a Proper Import Script

We don't like errors when running our imports and, unfortunately since we're human, we're going to have a lot of them. Instead of fixing things piecemeal, it's always better to just rerun everything.

Extraction: Summary

We did a lot in this section! We dipped our toes into the lovely world of SQL and we also learned that we should NEVER trust a spreadsheet!

Find the Flybys

Flybys: Introduction

We have our tools and our initial extraction ready to go, now let's get to work finding the flybys of Enceladus! In this section we'll focus on transforming this data into something we can query with some degree of confidence and then, hopefully, we'll find the exact time windows for the flybys.

Concept: Normalization

Structuring a relational database is (typically) all about following the rules of normalization, called "normal forms". Sounds theoretical, but it's pretty straightforward.

Narrowing Our Search

Before we can normalize the mission plan data, we need to understand what's in the table and how it's related.

Isolating the Enceladus Data

Our inspection showed us how we can isolate the Enceladus data - now let's do it!

Creating the Working Set

It's time to nail down the flybys! We have a plan table full of Enceladus mission plan goodness - now we just need to figure out how to sift the data correctly.

Summary of Flyby Work

We learned a few wonderful things in this section - one of the biggest, to me, is that our job is more than running queries - it's also digging in to the data we have and ensuring that it's usable.

INMS Data Import

Introducing the INMS

The fun begins! We have the data we need isolated and we feel good about its integrity - now we're ready to let it tell us a story. But what story is that? This is where things get interesting.

Extracting and Loading the INMS Data

We've done this all before so let's get rolling! The first thing we're going to do is to load up the INMS and chemical data into Postgres.

Transformation, Part 1

The INMS CSV is loaded, now we need to create our analysis table using the full power of PostgreSQL - specifically strong data types with appropriate constraints.

Concept: Constraints

We've been moving along at a pretty fast clip and it's crucial that we don't go too fast, skimming over super important concepts like constraints!

Transformation, Part 2

Now that we understand constraints a little bit more, let's carry on and buff this table out completely. We'll add a timestamp with time zone so we know when the import happened and then we'll get into some weird stuff with generated columns!


Wouldn't be any fun if we did everything right the first time, would it? Imposing rules during transformation is always problematic - which is great! We get to learn about our assumptions and also dig deeper into the data.

INMS Loading Summary

That was fun work! We learned a lot of fun concepts and some new SQL!


Understanding What We're Looking For

The fun begins! We have the data we need isolated and we feel good about its integrity - now we're ready to let it tell us a story. But what story is that? This is where things get interesting.

Concept: Joins

Bias can creep in anywhere during the analysis process - even in the way you structure your query and the joins you use.

Spreadsheet Export

Excel is EVERYWHERE and for good reason - it's simple to use, you can visualize data and you can even write simple functions and macros. Excel is the powerhouse of the analytical world and we need to prepare our data to work with it.

Ethical Considerations

It's important to take a step back from time to time and consider just what it is you're doing and why. In our case, we're working with one of the most important scientific data sets of the last century - at least with respect to planetary studies.

Ship It!

We feel good about the data and we're ready to ship it off! But how? Well that, friends, is the best part! There are multiple ways to do this but by far the simplest way is with a simple shell command using PSQL and \copy. You can drop the file anywhere you like - on your desktop so you can email it, or, what's easiest, is to use a secure file sharing system like Dropbox, Google Drive, OneDrive - whatever. Nice and fast.

Summary, and Farewell!

Parting... is such sweet sorry! I hope you were able to follow along with me in this section - because if you did your SQL and data skills just shot up - yay for you! Thanks for watching...

This is a premium course

And it's amazing and you'll love it, I promise (and guarantee it)

Find Out More

This course includes
  • Videos 28
  • Duration 2h 32m
  • Skills Beginner
  • Language English