The power of data and R

This week I started a data analysis MOOC!

Advertisements

Review of Module 1 of Data Analysis for Social Scientists (MITx, edX) – Introductory Lecture.

The introductory lecture of the course was really about showing how data could be visualised and interpreted to make (or be wary about making) conclusions about the world. Prof Duflo used a number of real life examples that demonstrated the power of data really well. I especially liked the network diagram, possibly because I’ve never had the opportunity to organise data in this way! Perhaps we should do one to map the volunteers at WISE. Would anybody like to take up the challenge? 😉

This module also included introductory exercises to R, which we will be using throughout the 12-week course. I self-taught myself R two years ago to analyse the survey data for my dissertation, but I had not used it since then. This course is a great opportunity to revisit, within a structured learning environment, quantitative data analysis methods and programming. I just heard a claim about how, in journalism, 90% of everybody forgets 90% of everything in 90 days (h/t Ezra Klein show with Ben Thompson). I think the same principle applies to any knowledge or skill we learn. Our brain get rusty if we don’t keep revising or practising!

The course will explore regression and econometrics, design of experiments, randomized control trials (and  A/B testing), machine learning, and data visualization. I’ve touched on these topics before (somewhere) but the topics now seem like a distant memory. I look forward to delving deeper in future weeks!

 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s