To Learn Data Science Better, Use SCIENCE!

Apr 16, 2018 | 1953 Views

If you are learning a new skill, think about HOW you are learning.

My plan was to never be a student again. I like learning. I enjoy the process of learning and as a teacher and scientist, I am continuously engaged in this process. I just thought I was done with the student role. I was wrong. I am back. This time, however, I am thinking about my process of learning in a new way in a scientific way.

Embracing the science of learning
People have spent a lot of effort on understanding how we learn. I find that keeping evidence-based learning methods in mind as I study data science helps me with coding in Python, building machine learning algorithms, and being the mentee instead of the mentor. The following three ideas help me stay focused and optimistic.

(1) Growth Mindset:
This year my kids came home from third grade with handouts to guide them toward growth mindsets. Their classrooms have bulletin boards dedicated to the concept, and they are taught that a fixed mindset is when you think that your intelligence, skills, and talents are fixed and will not change. In contrast, having a growth mindset means believing that traits can develop and change (Dweck 2006). By working, studying, and learning, one's intelligence, skills, and talent can develop. There is a lot out there about how to apply growth mindset in schools, for individuals, and even for organizations. My kids are taught that developing a growth mindset is as simple as saying I am not good at long division YET.

If you are trying something new, or if you took a break from school and now you are back, or if you are on a non-traditional trajectory into an analytical fieldâ?¦ you probably already have a growth mindset. If you already have a growth mindset and are learning new skills, I say seize the concept, embrace it, foster it; it may be one of the elements that helps you climb the data science mountain in front of you. As I worked through the initial exploratory data analysis stages of my first Capstone project for my data science program with Springboard, I channeled my inner third grader and reminded myself I haven't run my first machine learning algorithm YET.

(2) Active-learning:
Simply put, active learning means learning by doing. In college STEM (science, technology, engineering & mathematics) classes, active learning methods improve student learning (Freeman et al. 2014). Some educators intentionally incorporate active learning techniques. If you participated in a classroom debate, spoke up during a class discussion, were asked to think-pair-share with a neighbor during a lecture, designed and performed your own research experiment, or gave an in-class presentation, you have been an active learner. The contrast here is passive learning or â??sage-on-the-stageâ?? where, in traditional academic settings, you may have received all content through passive methods by listening to lectures. A lot of research has been done comparing active versus passive learning, including attempts to identify the best evidence-based approaches for effective teaching and learning (AAAS 2009, Derting et al. 2016).

If you are serious about becoming a data scientist, if you join a data science boot camp, if you study Python or R through interactive online tutorials where you enter code and run it (e.g., DataCamp, Mode Analytics, Codecademy) you WILL be gaining skills through active learning. You may encounter some online lectures that are purely passive learning - that is fine and does not mean you will not learn concepts and ideas. But as you study data science and you improve your coding and analytical skills, you will be learning those skills because you are practicing them actively. Take a moment to acknowledge the importance of each mini-project, late-night code debugging, hacking success and each piece of your Capstone project.

(3) Taking notes by hand:
This last learning concept is a simple one. Studies show that taking notes by hand, with pen and paper, increases the amount of information you learn and remember (Mueller & Oppenheimer 2014). Students performed better on conceptual assessments after taking handwritten notes compared to students who typed their notes. Apparently, the connection you make with the concept is more concrete when you write it down than if you type your notes during a lecture. When I teach college classes, I describe this study on the first day, but I do not require that students hand-write their notes. I think students might have their own reason for typing notes and that it is a personal preference.

Conclusion:
Now that I am back in the student seat, I have filled several notebooks with messy, busy notes on the key differences between Support Vector Machines and K-Nearest Neighbors classifiers, on methods in natural language processing, on how to approach an informational interview. I am learning by writing it down. I am learning by doing. I am learning all about data science because I believe I can develop new skills. Whatever the next big topic in my course is: I have not mastered that type of analysis YET.

Sources:
AAAS. 2009. Vision and Change in Undergraduate Biology Education: A Call to Action. Washington DC: American Association for the Advancement of Science.

Derting, T.L., Ebert-May, D., Henkel, T.P., Maher, J.M., Arnold, B., and Passmore, H.A. 2016. Assessing faculty professional development in STEM higher education: Sustainability of outcomes. Science Advances, 2(3), e1501422e1501422. http://doi.org/10.1126/sciadv.1501422

Dweck, Carol. 2006. Mindset: The New Psychology of Success.

Freeman, S., Eddy, S.L., McDonough, M., Smith, M.K., Okoroafor, N., Jordt, H., and Wenderoth, M.P. 2014. Active learning increases student performance in science, engineering, and mathematics. Proceedings of the National Academy of Sciences of the United States of America, 111(23), 84108415.
http://doi.org/10.1073/pnas.1319030111

Mueller, P.A. , Oppenheimer, D.M. . 2014. The Pen Is Mightier Than the Keyboard: Advantages of Longhand Over Laptop Note Taking, Psychological Science, Vol 25, Issue 6, pp. 11591168.
https://doi.org/10.1177/0956797614524581



The article was originally published here

Source: HOB