Learning Data Science as a Community
Organizing a 20-day campaign revolved around learning and teaching Data Science
Being a beginner in data science does pose obvious challenges such as being overwhelmed by the sheer expanse of topics to cover, the ever-persistent impostor syndrome, spurious knowledge and confusion induced by online articles.
A good way to tackle these challenges could be to start learning together as a community, guided by steps figured out by fellow beginners and experts alike. This was our focus when we designed the 20 day #DataDecember initiative at the London School of Economics and Political Science Data Science Society.
The #DataDecember Initiative was a month long social media campaign to provide data science beginners a platform to learn core data concepts in an easy-to-understand and useful way. It ran throughout the entirety of December and can now be accessed on the #DataDecember Official Website.
If you are a beginner, this is all you need to get a solid start into the field. And if you are interested in how we led this community-driven learning initiative, do read on. There are also tips on how you could hold one in your institution or workplace to learn together!
The main idea was to not add onto the monotony of data science resources already prevalent on the Internet. We wanted to work on something that would:
- Help us learn
- Help others learn
- Not waste anybody’s time
- Be driven by value than by emotion
This led us to clearly defining a plan that involved releasing short articles (under 500 words) per day on a particular concept X and supplement the material with good references(preferably what we have used to learn concept X).
Good plan? Maybe not.
This initial plan lacked an important component — Empathy. A set of short articles released on a daily basis cannot be of help to complete beginners unless the articles have a way of connecting with these beginners.
This is where we chose to introduce the section “What the beginner thinks”. The notion was to write in this section our thoughts we had when we first encountered the topic of the article. In all honesty, writing this section helped us really understand the barriers a beginner could have when faced with the task of learning about concept X.
Once the initial idea was developed, the next task was to put this all down as a process. There had to be a specific set of steps that we had to take if we were to make this work.
A basic checklist was developed that entailed the following TODOs
- Create a team to work on this
- Make a list of concepts we should cover
- Create a webpage using Github pages to organize all content
- Decide on the social media strategy
- Write articles
Each of these steps had their own subsequent smaller steps, but I shall spare you the details!
In terms of the content, we placed our focus on organizing the content on the basis of 4 parts of Data Science , one for each week— Data Science Fundamentals, Stats and Math for Data Science, Computer Programming for Data Science and Data Science Project Week.
Data Science Project Week was built on the idea of explaining one facet of a complete data science project in high-level fashion. It covered the ideas of Problem Formulation and Data Acquisition, Data Cleaning, Data Exploration, Modelling and Presentation.
Each article was capped at less than 500 words and had to have at least 2 good references that could help readers dive in deeper. Also, we made it a point to include one quote per article as inspiration is always very important to learn anything.
We had two teams working on making #DataDecember useful to our readers at LSE. One team was the team that wrote down the articles and updated it on the website daily. The other was entrusted with the responsibility of publishing graphics and promo material of the daily content on our social media channels.
In a nutshell, #DataDecember was the combined effort of 14 full-time university students driven by a passion to write as we learn. And more importantly, write as we would speak to someone looking for data science advice.
As Rachel Thomas, co-founder of fast.ai puts it, “You are best positioned to help people one step behind you”. The #DataDecember was our way of believing in that notion and giving back to the constantly evolving and ever-energetic data science community!
What we learnt through organizing #DataDecember
#DataDecember definitely did not become the next biggest hashtag of the internet, contrary to our expectations.
However, that does not affect us because it provided us with something else — An opportunity to revisit the fundamentals of data science as beginners.
Personally, #DataDecember helped me hone my abilities to co-ordinate an initiative which was larger than any personal learning journey I have ever taken up.
As a student club, we came closer with this initiative and hopefully, managed to inspire our members in the process.
If you want to know about further cool things we are working on at the society follows us on
Want to conduct a similar campaign at your university or workplace?
If this sounds like you, and if you would like a more detailed understanding of all the planning procedures involved to run a campaign like this, feel free to reach out to me at firstname.lastname@example.org.