If you’re interested in data analysis and data visualization, you’re probably wondering how to get started. What’s the best way to add data analysis and data visualization skills to your portfolio? We believe the topic is so important we’ve created a page that talks about career and organizational benefits and how to get started on or uplevel your data science skills.
Data science is a large and expanding career field, with many different data science jobs, and there are lots of ways to obtain data analysis and data visualization skills. You must decide if you want to become a data scientist or if you’d rather add data analysis and visualization skills to your current job role. Either way, in this article, we’re going to look at what it takes to get started on your data science journey. From the skills you need to what you must have on your resume and the different training and education options you can choose, and the benefits of each.
Education
Traditionally, entry-level data scientist positions start with masters or Ph.D. requirements. A master's degree typically requires 30-36 credits, and while it can be completed in a single year, two years is more common, especially for people who are working and going to school part-time. The cost of a masters is around $30,000. A Ph.D. requires 90-120 credits and must be completed within seven years, and can be completed in two years (if you don’t like the company of humans or seeing the sun) and is often completed in four and a half to five years. Ph.D. programs typically cost $30,000 a year, so the total cost can run upwards of $150,000.
Those types of programs are certainly valuable, and most postgraduate programs facilitate connections to local employers as part of the coursework and offer job placement assistance after graduation. These choices often come down to your current life stage, commitments, and drive. While financial assistance is relatively easy to get, you will have to pay it back, and it doesn’t help pay for a mortgage or children if you have either of both of those. Like an MBA, a master's degree in data science is often a better choice for people working a full-time job and is still a very large time and monetary commitment. Beyond those types of programs, you have options like massive online open courses (MOOCs) covered by Udacity, Courser, or others, which can give you a nano degree. Those programs offer a lot of different courses and allow you to go fairly deep into many areas of data science, but the learning is self-paced. That is to say, it's up to you. Many programs do offer mentors that are available to give advice, support, and review projects.
Outside of that, you have the option of teaching yourself data science with online self-paced learning. That approach will be entirely dependent on your ability to stay motivated and put in the time, the same as with a nano degree program, but you won’t likely have access to a mentor unless you know a data scientist personally who’s willing to take on that role. You might also struggle to find good sample data science projects or data to play and learn with. Then there’s instructor-led training like we offer, which puts you in daily contact with your instructor, who is with you every step of the way. While you leave those courses with a very defined set of skills that directly translate to job tasks, these courses require your focused attention for a few days or a few weeks, depending on what you’re trying to learn.
With the options laid out, let’s look at what employers look for when they hire for data science positions.
What Employers are Looking For
Employers generally list a masters degree as a requirement for entry-level positions, but many also will accept candidates with equivalent experience. What is equivalent experience? That’s up to the employer. Organizations use data science in many different ways. Manufacturers may use it to optimize processes, retailers may use it to anticipate inventory levels, logistical companies to determine population movement, and healthcare companies to identify deformed human cells. The possibilities are endless, and an article from KDNuggets that examined data science position hiring practices stated:
"Every company looks for something different. What gets you hired at Google may or may not work at other companies (and may even be meaningless). So building the “perfect” all-purpose data science resume is all-but-impossible.”
Tip. If you’re applying for a data science job, ask as many questions as possible of the recruiter prior to the interview to determine how data science skills are used, what business or organizations goals data science is helping achieve, what type of data you’re working with, how AI and machine learning are used, what data pipelines are in place, and what estimations need to be made.
In that same article, they did list some issues that will get your resume discarded by recruiters. Those include:
Focusing on experience with projects from known “training-wheels” data sets from nano-degree programs, including the survival classification on the Titanic data set, the hand-written digit classification on the MNIST data set, and the flower species classification using the iris data set. Recruiters are familiar with these datasets and might not take you seriously if that’s all they see on your resume.
Having only MOOCs projects in your portfolio.
Focusing on areas of data science that are already heavily explored rather than having new work.
In short, if all you’ve done is classroom work with well-known training data sets, you’re not going to stand out from the pack to recruiters or prospective employers. That said, if you’ve done something special, such as a deep dive or performed sophisticated analysis with a training data set, that may be worthy of space on your resume if it’s relevant to the work being done by the organization you’re applying to. Remember, employers, want the best people, and recruiters are frequently charged with finding the top 1% or 2% of candidates. You have to show that you’re special in your field to stand out.
Experience is Key
You have to show experience, and you have to show experience beyond coursework and training data sets. So how do you get that? You make data science your hobby and build up a portfolio of projects that increase your knowledge, showcase your skill, and help you stand apart to recruiters reviewing your resume and in interviews. Here are a few tips:
Analyze publicly available data sets to find interesting information.
Compete in Kaggle competitions.
Read everything you can about data science trends and findings, and replicate analysis.
Tip: Make sure you acquire other must-have skills used in almost all data science projects, including the use of version control systems like GitHub, DevOps for developing and deploying data science solutions and experience with DevOps platforms and tools like GitLab, Python, and Flask, etc., and experience in working with databases such as SQL, MySQL, MongoDB, and others.
The Value of Data Science Training
So if you need experience beyond what’s delivered in training, you’re likely asking what’s the value of training? The true value of training is the acquisition of skills. From a degree program to MOOC, to self-paced, to instructor-led professional training, the true value you get is the skills you walk away with. You can’t analyze any public data sets or compete in Kaggle competitions if you don’t know how to perform data analysis to begin with. So training provides the fundamental stills, the foundational structure on which to build your data science career. Given that, you have more information to consider your training options.
The price of training and the time it takes to complete it are huge factors in choosing how to acquire your data science skills. One reason we have chosen to deliver immersive instructor lead training is that it takes a relatively small amount of time, a few days to a week per course, but during the course, you’re focused on the training, and you have an instructor that is there to provide guidance, insights, and real-world examples every minute of the course and we also offer office hours and email support for a period following the course. We feel deep immersion, focused attention, and lots of practice is the best way for adults to acquire new skills.
The type of training you take can also provide additional value. For example, our courses not only teach skills, but also tools and languages that are in demand, such as Tableau, Power BI, and Python. You can apply those skills in a work environment the day after class or add them as separate skills on your resume. That can help you get a job and get paid, or get paid more in your current job, as you’re progressing on your data science career path.
Choosing the Best Path
We’ve covered what it takes to start a data science career path. The first step is acquiring fundamental data analysis skills. How you do that depends on the time you have, the cost you’re willing to pay, and how you learn best. You should also factor in additional benefits training can provide, such as job placement from college degree programs or in-demand software skills from immersive instructor-led training courses. In the end, the decision is yours, and you should accept the fact that your journey to becoming a data scientist is likely to be a long one. You should select the training that helps you advance on your career path, works with everything else in your life, and, hopefully, provides additional benefits in the process.
If you’re still considering if these skills can help you, your career, or your organization see our Data Analysis and Visualization for Career Growth page.
