*hacker voice* “I’m in” – Coding and Software for Data Analysis

While data analysis has existed in one form or another for centuries, its modern concept is highly tied to a digital environment, which means that people who are looking to move into the data science field will undoubtedly need some technology skills. In the data field, the primary coding languages include Python, R, and SQL. Software is a bit more complicated, with numerous different programs and services used depending on the situation, including Power BI, Spark, SAS, Excel, to name a few. While this is overwhelming, remember that it is not important to become an expert in all of the languages and software. Becoming skilled in one language and a few of the software options, depending on your interest or on the in-demand skills on job listings, will give you the transferable skills to quickly pick up the other languages and software as needed. If this still seems to be  an overwhelming prospect, remember that the best way to eat an elephant is one bite at a time. Take your time, break up the task, and focus on one step at a time! 

LinkedIn Learning

  1. Python for Data Science Essential Training Part 1 
    1.  This 6 hour course guides users through an entire data science project that includes web scrapers, data cleaning and reformatting, generate visualizations, preform simple data analysis and create interactive graphs. The project will have users coding in Python with confidence and give learners a foundation in the Plotly library. Once completed, learners will be able to design and run their own data science projects.  
  1. R for Excel Users 
    1. With Excel being a familiar platform for many interested in data, it is an ideal bridge to more technical skills, like coding in the R language. This course is specifically designed for data analytics with its focus on statistical tasks and operations. It will take user’s Excel skills to another level while also laying a solid foundation for their new R skills. Users will be able to switch between Excel and the R Desctools package to complete tasks seamlessly, using the best of each software to calculate descriptive statistics, run bivariate analyses, and more. This course is for people who are truly proficient in Excel but new to R, so if you need to brush up your Excel skills, go back to the first post in this series and go over the Excel resources!   
  1. SQL Essential Training 
    1. SQL is the language of relational databases, so it is of interest to anyone looking to expand their data handling skills. This training is designed to give data wranglers the tools they need to use SQL effectively using the SQLiteStudio Software. Learners will soon be able to create tables, define relationships, manipulate strings, use triggers to automate actions, and use sub selects and views. Real world examples are used throughout the course and learners will finish the course by building their own SQL application. If you want a gentler introduction to SQL, check out our earlier post on SQL Murder Mystery  

O’Reilly Books and Videos (Make sure to follow these instructions for logging in!) 

  1. Data Analysts Toolbox – Excel, Python, Power BI, Alteryx, Qlik Sense, R, Tableau 
    1. This 46 hour course is not for the faint of heart, but by the end, users will be a Swiss army knife data analyst. This isn’t for true beginners, but rather people who are already familiar with the basic data analysis concepts and have a good grasp of Excel. It is included in this list because it is a great source for learning the basics of the myriad of software and programming languages that data analysts are expected to know, all in one place. The course starts with teaching users about advanced pivot tables, so if users have already mastered the basic pivot table, they should be ready for this course.  
  1. Programming for Data Science: Beginner to Intermediate 
    1. This is an expert curated playlist of courses and book chapters that is designed to help people who are familiar with the math side of data analysis, but not the computer science side. This playlist gives users an introduction to NumPy, Pandas, Python, Spark and other technical data skills. Some previous experience with coding may be helpful in this course, but patience will make up for lack of experience.  

In the Catalog

  1. Python crash course : a hands-on, project-based introduction to programming 
    1. Python is often lauded as one of the most approachable coding languages to learn and its functionality makes it popular in the data science field. So it is no surprise that there are a lot of resources on and off campus for learning Python. This approachable guide is just one of the many resources available to UIUC students, but it stands out with its contents and overall outcomes. “Python Crash Course” covers general programming concepts, Python fundamentals, and problem solving. Unlike some other resources, this guide focuses on many of Python’s uses, not just its data analytics capabilities, which can be appealing to people who want to be more versatile with their skills. However, it is the three projects that make this resource stand out from the rest. Readers will be guided in how to create a simple video game, use data visualization techniques to make graphs and charts, and build an interactive web application.  
  1. The Book of R : a first course in programming and statistics 
    1. R is the most popular coding language for statistical analysis, so it’s clearly important for data analysts to learn. The Book of R is a comprehensive and beginner friendly guide designed for readers who have no previous programming experience or a shaky mathematical foundation as readers will learn both concurrently through the book’s lessons. Starting with writing simple programs and data handling skills, learners will then move forward to producing statistical summaries of data, preforming statistical tests and modeling, create visualizations with contributed packages like ggplot2 and ggvis, write data frames, create functions, and use variables, statements, and loops; statistical concepts like exploratory data analysis, probabilities, hypothesis tests, and regression modeling, and how to execute them in R; how to access R’s thousands of functions, libraries, and data sets; how to draw valid and useful conclusions from your data; and how to create publication-quality graphics of your results.  

Join us next week for our final installment of the Winter Break Data Analysis series: “You can’t analyze data if you ain’t cute: Data Visualization for Data Analysis”    

Learn Data Analysis: What’s Math Got to do With It?

What’s math got to do with data analysis? Unfortunately, for those of us who are chronic humanities people, math has a lot to do with it. This might seem like a daunting barrier, especially if the last time you looked at a math problem was in a high school algebra class. This is also true for learners who are already skilled with the technological aspect of data analysis but are not familiar with the mathematics side of thing. However, there are so many resources available to help self-directed students learn the basics and get up to speed for the purposes of data analytics! Using the resource platforms described in last week’s blog post, these resources will have even chronic humanities people playing with numbers in no time!  

LinkedIn Learning 

  • Learning Everyday Math 
    • Look, some of us did not absorb or retain the basic math lessons of our early education. That’s okay! This is a no-judgment zone, and this 2 hour course will help users learn how to calculate percentages for tips and taxes, compare prices while shopping, find the area and volume for home-improvement projects, and learn the basics of probability.  
  • Become a Data Scientist 
    • This 21 hour Learning Path is made up of 12 courses that focus more on the statistical side of data analysis than the technical steps of the process. This course is more geared toward users with experience in IT and computers, so it is not the best for people who do not have a strong technical background. However, for those who are familiar with computer science and want to pivot into data analytics, this is an ideal curriculum.   

O’Reilly Books and Videos (Make sure to follow these instructions for logging in!)

  • Essential Math for Data Science 
    • This eBook mixes basic coding skills with math lessons to cover the essential analytical skills needed for data science work. Relevant aspects of calculus, probability, linear algebra, and statistics and how they apply to techniques like linear regression, logistic regression, and neural networks are covered in plain English. The chapters include exercises with answers for self-assessment as well as career advice for budding data analysts.   
  • Statistics for Data Science using Python 
    • Besides books, O’Reilly also has expert curated playlists that consist of chapters of several different books, videos and more. This is a great way of getting the most out of several resources to focus on a single skill. This playlist covers the essential statistic concepts found in 11 different resources. Learn about Normal distribution, hypothesis tests, p-values, central limit theorem and more without having to dig for the resources yourself!  
  •   Data Science 101: Methodology, Python, and Essential Math 
    • On top of books and playlists, O’Reilly also has video-based courses. This course covers a lot of data analytics basics, but those who want to focus on the math aspect will benefit from Chapters 15-19. These chapters cover linear algebra, mathematical structures, probability, random variables and multiple variables, and statistical inference.  

In the Catalog 

Be sure to come back next week for the thrilling continuation with “*hacker voice* I’m In: Coding and Software for Data Analysis! 

Learn Data Analysis Over the Winter Break!

In the last twenty years, humanity has become super proficient in collecting data. Therefore, It is no surprise that the skills to analyze that massive collection of data is in ever increasing demand on the job market. For those of us who are worried about future job prospects, learning these in-demand data analysis skills seems like a logical next step, even if they do not fit into our current degree program. Fortunately, the university has a plethora of self-guided resources available for students looking to build their data skills. What better time to use these resources than during the long winter break?! Over the next few weeks, this blog will delve into the available resources that cover the three main skill areas of data analysis: math, coding and software, and visualization. 

Before diving into those areas, it is wise briefly look at the foundations of data analysis as well as the resources that will be showcased this month. Take this week to get acquainted with these different resource platforms and learn a few starting skills! 

LinkedIn Learning

All UIUC students have access to LinkedIn Learning. Simply login with your NetID credentials, just be sure you are logging into LinkedIn Learning, not the main LinkedIn site.  You will have access to a whole trove of high-quality videos and courses designed to help you learn career-building skills. Not only are the videos professional grade, but they often have accompanying exercise files, learning groups, certificates and exams. The collection ranging from short 5-15 minute videos that teach specific function or skills to dozen hours long courses that are designed to give a comprehensive foundation. The best part of using LinkedIn Learning is that the course and certificates completed here are then displayed on personal LinkedIn pages, showing potential employers that users have the skills they are looking for. 

  • Data Analytics for Students
    • This course is for the true data analytics babies out there. This introduction gives users the basic understanding of what data analytics is, the skills users will need to be successful,  the software and tools common in the field and what careers in data analytics look like. This 1 hour course is well worth the time for those who aren’t sure where to start their data journey.
  • Career Essentials in Data Analysis by Microsoft and LinkedIn
    • Discover the skills needed for a career in data analysis. Learn foundational concepts used in data analysis and practice using software tools for data analytics and data visualization. This is a Learning Path made up of 3 different courses that has about 9 hours of content for students to work through on their own schedule. The courses have exams for self-evaluation as well as a final exam that earns users a professional certificate. 
  • Excel: Managing and Analyzing Data
    • We have all put “proficient in Excel” on a resume, but wouldn’t it be nice if that was actually true? Unlike other data analytics courses, this course focuses on one program that most modern users are already familiar with but do not truly harness the power of. This is ideal for baby data analysts as it doesn’t bombard learners with a whole new software ecosystem but still teaches the transferable skills all data analysts use. Running at just under 4 hours, this course efficiently and comprehensively teaches users impressive data analytics skills. 

O’Reilly Books and Videos

This is a lesser known resource available at UIUC but it has some great online books and videos that tend to focus on the scientific and technical fields. Logging in is not straightforward, unfortunately. The best way to get there is to go to the Library Catalog’s record for a book offered through O’Reilly (Like this book on Python) and then follow the instructions on this LibGuide to log in. Once you are in, you will see a sizable collection of e-books and courses. The materials skew towards the more experienced users, but there are a few resources that will help baby data folks really develop their skills. 

Library Catalog

Learn data science the old fashion way, with books! There are a lot of books available at UIUC libraries for students who want to teach themselves a new skill. Here are a few choices for people looking for an easy introduction to data analysis. The Scholarly Commons collection is easily accessible and found just to the right of the main entrance to the stacks. 

Be sure to check back here next week for our next installment, “What’s Math got to do with it?”!

Making Infographics in Canva: a Guide and Review

Introduction

If you’ve ever had to design a poster for class, you’re probably familiar with Canva. This online and app-based graphic design tool, with free and subscription-based versions, features a large selection of templates and stock graphics that make it pretty easy to create decent-looking infographics. While it is far from perfect, the ease of use makes Canva worth trying out if you want to add a bit of color and fun to your data presentation.

Getting Started

Starting with a blank document can be intimidating, especially for someone without any graphic design experience. Luckily, Canva has a bunch of templates to help you get started.

Canva infographic templates

I recommend picking a template based on the color scheme and general aesthetic. It’s unlikely you’ll find a template that looks exactly how you want, so you can think of a template as a selection of colors, fonts, and graphics to use in your design, rather than something to just copy and paste things into. For example, see the image below – I recently used the template on the left to create the infographic on the right.

An infographic template compared to the resulting infographic

General Design Principles

Before you get started on your infographic, it’s important to remember some general design guidelines:

  1. Contrast. High levels of contrast between your background and foreground help keep everything legible.
  2. Simplicity. Too many different colors and fonts can be an eyesore. Stick to no more than two fonts at a time.
  3. Space. Leave whitespace to keep things from looking cluttered.
  4. Alignment and balance. People generally enjoy looking at things that are lined up neatly and don’t have too much visual weight on one side or another.
An exaggerated example of a design that ignores the above advice.

Adding Graphs and Graphics

Now that you have a template in hand and graphic design principles in mind, you can start actually creating your infographic. Under “Elements,” Canva includes several types of basic charts. Once you’ve added a chart to your graphic, you can edit the data associated with the chart directly in the provided spreadsheet, by uploading a csv file, or by linking to a google spreadsheet.

Canva interface for creating charts

The settings tab allows you to decide whether you want the chart to include a legend or labels. The options bar at the top allows for further customization of colors and bar or dot appearance. Finally, adding a few simple graphics from Canva’s library such as shapes and icons can make your infographic more interesting. 

Examples of charts available in Canva, with a variety of customizations.

Limitations and Frustrations

The main downsides to Canva are the number of features locked behind a paywall and the inability to see only the free options. Elements cannot be filtered by price and it seems that more and more graphics are being claimed by Canva Pro, so searching for graphics can be frustrating. Templates can be filtered, but it will still bring up results where the template itself is free, but there are paid elements within the template. So, you might choose a template based on a graphic that you really like, only to find out that you need a Canva Pro subscription to include that graphic.

The charts in Canva also have limitations. Pie charts do not allow for the selection of colors for each individual slice; you have to pick one color, and Canva will generate the rest. However, if you want to have more control over your charts, or wish to include more complicated data representations, you can upload charts to Canva, which even supports transparency.

Conclusion

As mentioned above, Canva has its downsides. However, Canva’s templates, graphics, and charts still make it a super useful tool for creating infographics that are visually appealing. Try it out the next time you need to present some data!

There’s been a Murder in SQL City!

by Libby Cave
Detective faces board with files, a map and pictures connected with red string.

If you are interested in data or relational databases, then you have heard of SQL. SQL, or Structured Query Language, is designed to handle structured data in order to assist in data query, data manipulation, data definition and data access control. It is a very user-friendly language to learn with a simple code structure and minimal use of special characters. Because of this, SQL is the industry standard for database management, and this is reflected in the job market as there is a strong demand for employees with SQL skills.  

Enter SQL Murder Mystery

In an effort to promote the learning of this valuable language, Knight Labs, a specialized subsidiary of Northwestern University, created SQL Murder Mystery. Combining the known benefits of gamification and the popularity of whodunit detective work, SQL Murder Mystery aims to help SQL beginners become familiar with the language and have some fun with a normally dry subject. Players take on the role of a gumshoe detective tasked with solving a murder. The problem is you have misplaced the crime scene report and you now must dive into the police department’s database to find the clues. For true beginners with no experience, the website provides a walkthrough to help get players started. More experienced learners can jump right in and practice their skills. 

I’m on the case!

I have no experience with SQL but I am interested in database design and information retrieval, so I knew it was high time that I learn the basics. As a fan of both games and detective stories, SQL Murder Mystery seemed like a great place to start. Since I am a true beginner, I started with the walkthrough. As promised on the website, this walkthrough did not give me a complete, exhaustive introduction to SQL as a language, but instead gave me the tools needed to get started on the case. SQL as a language, relational databases and Entity Relationship Diagrams (ERD) were briefly explained in an approachable manner. In the walk through, I was introduced to vital SQL functions like “Select:, “Where”, wildcards, and “Between”. My one issue with the game was in the joining tables section. I learned later that the reason I was having issues was due to the tables each having columns with the same title, which is apparently a foundational SQL feature. The guide did not explain that this could be an issue and I had to do some digging on my own to find out how to fix it. It seems like the walkthrough should have anticipated this issue and mentioned it. That aside, By the end of the walkthrough, I could join tables, search for partial information matches, and search within ranges. With some common sense, the database’s ERD, and the new SQL coding skills, I was able to solve the crime! If users weren’t challenged enough with that task, there is an additional challenge that suggests users find the accomplice while only using 2 queries.

User interface of SQL Murder Mystery
Example of SQL Murder Mystery user interface

The Verdict is In

I really loved this game! It served as a great introduction to a language I had never used before but still managed to be really engaging. It reminded me of those escape room mystery boxes like Hunt a Killer that has users solve puzzles to get to a larger final solution. Anyone who loves logic puzzles or mysteries will enjoy this game, even if they have no experience with or even interest in coding or databases.  If you have some free time and a desire to explore a new skill, you should absolutely give SQL Murder Mystery a try!