Learn Data Science (or any skills) with "Deliberate Practice"

Author:Murphy | View: 27943 | Time: 2025-03-23 13:05:51

Recently I read a book named Peak: Secrets from the New Science of Expertise written __ by Anders Ericsson and Robert Pool. The book challenges the common myth that expertise comes from natural talent. Rather, with ample examples in different domains, it proves top performances can be achieved, and skills can be learned through such a technique called "Deliberate Practice." In the book, the author provides examples of various skills, such as excelling in musical performances, wartime aircraft battles, and strategy games like chess and Go. Reading this book brings me a deja vu as I reflected on my own journey of learning Data Science skills fast towards mastering.

In my previous articles, I shared how did I become a data scientist:

How I became a data scientist

and the seven principles I follow to be a better data scientist:

Seven Principles I Follow To Be a Better Data Scientist

In this article, I want to share the techniques discussed in Peak as Deliberate Practice and illustrate them with my own learning journey. Hope it will be helpful for those who aim to learn data science or any other skills.

Deliberate Practice vs. Naive Practice

First thing first, what is Deliberate Practice? In contrast to naive practice, which is essentially doing something repeatedly and hoping the repetition itself will do the job. For example, play the same song every day with a guitar to be a guitar guru, or run the same trail every day at the same speed to prepare for a marathon. Indeed, repetition brings familiarity. You might become an expert in playing one song and lose some weight through the daily run. However, expertise in one song or trail won't make you a top guitar performer or runner. No doubt we have hobbies and just want to do something for fun rather than for competition or make a living out of it, but for the skills you do desire to get better and eventually become a top performer, deliberate practice is a better way to achieve your goals. Deliberate practice is being purposeful, thoughtful, and focused during practice. Specifically, it has the following characteristics:

Deliberate practice has well-defined, specific goals for each stage. For example, if your goal is to be a top runner in a marathon, you need to have specific goals in how fast you should finish a marathon and make plans to train towards reaching it.
Deliberate practice is focused. During practice, you must be aware of what you do with full attention, not in auto-pilot mode. If you find yourself distracted while completing a task, you are likely not taking the goal seriously, or the tasks are too easy to fulfill. Working on easy tasks will not help you grow your skills.
Deliberate practice involves feedback. You have to be evaluated constantly for your performance to find areas of improvement. In most cases, you must follow mentors who will provide instant feedback at your practice. A mentor doesn't necessarily have to be a person. It can also be a well-designed tutorial when finding a mentor is impossible. As an illustration, the book explains how skilled writers enhance their writing abilities by contrasting their articles with the publications in which they aspire to feature their work. In addition, it is important to switch mentors as you continue to enhance your skills. For instance, after a writer's article was accepted by a desired publication, they altered their approach to practice when they set goals to improve vocabulary, conciseness, logical structure, etc.
Deliberate practice requires you to step out of your comfort zone. Unless your goal is to maintain your current skill level, if you want to get better from every practice, you should make yourself feel a bit "uncomfortable" during practice. Always challenge yourself to run faster, perform better, write that piece of code cleaner, and execute faster. Stepping outside your comfort zone extends your frontiers and opens up opportunities.

However, deliberate practice might not be applicable in all fields. As discussed above, the field has to have clearly defined and measurable criteria for what makes a top performer in this skill.

The book gives plenty of examples of how top performers from different fields have reached their levels through deliberate practice. In this blog post, I would like to share an example of how I successfully developed my data science skills from scratch.

I have specific and well-defined goals

Setting clear goals was the first step in my data science journey. Instead of aiming to "learn data science," I have always established specific objectives, such as:

Mastering Python programming so I can run data analysis for my research projects more effectively;
Understanding statistical concepts and Machine Learning algorithms so I know why some models work better in some scenarios and which model I should choose for my research projects;
Build a data science portfolio, so I can get interviews more easily.
Learn these models so I can use them to solve problems at work.

These goals served as guideposts, helping me navigate the vast data science landscape effectively, and they keep evolving overtime as I am facing bigger challenges, and at different settings.

Advice: It is important to set up goals before starting. Knowing what you want to do with the skills will guide you on what to learn and how to learn. If you are a student, you may learn something to pass an exam; If you are at work, you may learn something to equip yourself for a project you are working on or stay cutting-edge for the latest technology in your industry. We need both short-term goals and long-term goals. Anyone familiar with time management would know Eisenhower's Urgent/Important Principle concept. We are categorizing tasks according to their level of urgency and importance. Although many prioritize urgent and important tasks, it's important not to neglect important but non-urgent tasks that set us apart in the long term. For instance, learning and mastering a new skill. Short-term goals are like these urgent and important tasks, which give us efficiency, but long-term goals motivate us to grow continuously and consistently, which is especially valuable in the ever-changing world.

I stay focused during practice

With well-defined goals in place, staying focused during practice became essential. Instead of mindlessly consuming tutorials and resources, I structured my learning sessions. I dedicated focused blocks of time daily while taking online courses at Coursera. For these online classes, it is very easy to think you understand something because you watched the videos or read the articles. What I forced myself to do was to work on the projects to get my hands dirty. At first, I would spend hours googling different syntaxes and try them out on a DataFrame to illustrate their usage. A simple visualization would take me a long time. However, you can't learn a Programming language simply by copying and pasting. We need to know what it is used for and why you use this one over the others for specific tasks. Getting my hands dirty on projects helped me stay focused and absorb information deeply.

Advice: When practicing, it's important to bring your brain along. While it may seem obvious, don't assume that you're automatically staying focused. Staying focused requires more energy than simply going through the motions. It's similar to running – maintaining a certain heart rate while running a mile requires more energy than just running mindlessly. During practice, try to do more thinking and question yourself more often. Ask yourself if you truly understand what you're doing and why you're doing it.

I ask for feedback from practice

In several of my past blog posts, I've emphasized the significance of attending a data science bootcamp in my journey.

How to Benefit From Attending a Data Science Bootcamp?

The feedback I received from knowledgeable mentors was invaluable. Despite meeting virtually, I received feedback regularly on my lecture questions, homework assignments, and projects. In addition, my small group of classmates with diverse backgrounds provided feedback to each other. We learned from one another's expertise and mistakes. Their insights and critiques helped me refine my technical skills and inspired me to approach problems strategically and creatively. Although when taking online tutorials, I received grades for completing online assignments, I struggled without guidance and immediate feedback. Constructive feedback transformed my errors into opportunities for growth and accelerated my learning curve. Thanks to this intensive 8-week bootcamp, I grew exponentially.

Advice: It's important to have both time to practice and reflect on your own, but also to seek guidance and feedback from experienced individuals. This is the quickest and most effective way to improve your skills. When learning something new, consider finding a mentor who is much more sophisticated than you so can receive immediate feedback on your performance during practice. If you can't find a mentor, it is also useful to have practice partners, or join a community.

I step out of my comfort zone

Although the comfort zone can provide a sense of security, true growth can only be achieved by stepping beyond its boundaries. For me, learning data science was a journey outside of my comfort zone. In the field of Economics, Stata is a familiar software used alongside R by faculties and students for data analysis and modeling. However, with the increasing popularity of big data, I recognized the need to learn Python to improve my productivity. To truly master data science, I challenged myself with projects that seemed difficult at first. This allowed me to navigate through challenges and improve my problem-solving skills. Data science is changing so fast. What was new and cutting-edge last year can easily be replaced and become legacies. We can never be complacent in this field. Stepping out of my comfort zone not only expand my skill sets, but also equip me with the courage to accept challenges and confidence that I can eventually solve the problems I am facing.

Advice: Each trial outside your comfort zone expanded your skill sets and boosted your confidence. Don't be afraid of facing challenges during practice. Challenge yourself to work on more complicated projects, write cleaner code, and build even higher accurate or more explainable models. It is the most painful but also the most rewarding part of growing. Walking uphills is always more tiring than walking on the Plain in circles. You can always lie to yourself that walking in circles is also practice, but it will not lead you to the rewarding scenery on the top of a mountain.

In essence, deliberate practice is a purposeful and structured approach to skill development. It involves setting specific goals, maintaining focused practice sessions, actively seeking and incorporating feedback from mentors, and challenging oneself by stepping beyond the comfort zone. This method of practice transforms learning from a passive process into a proactive journey of growth and mastery. In this post, I share my experience of learning data science skills to further prove the importance of deliberate practice, hoping to inspire anyone who is interested in mastering any new skills.

Thanks for reading! Lastly, don't forget to: