Multiple Slack channels were created to support communication:
homework
social
Participants were provided with clear instructions on how to download and access Slack, but it turned out that only two thirds (of the 15 participants) decided to pursue that line of support middle east rcs data and join up. This demonstrates that intensive data -oriented online learning is likely better suited to the more technophilic amongst us.
Image of a Slack window showing participants who used the tool to collaborate.
Image: Collaborating on Slack
Syllabus highlights
The first day covered more theoretical and practice issues around the ethical and legal sourcing of data, such as from APIs, and keeping code documented whilst days 2-3 ramped up the tech, learning how to use various R packages to clean manipulate, join and map data, such as TidyR and Tidyverse, GGPlot, RLeaflet, Plotly and html widgets.
I’ve long been a great fan of the growing reproducibility agenda and have blogged on the challenges of this for qualitative data and ran the 14 February #LoveYourCode event with the ONS this winter.
With reference to this Big Data course, the theme of reproducibility ran throughout the week.
From day 1 we ensured participants were fully aware of the need to document, track and share code and properly cite any data they use.
We showed them how to use R Markdown and how to add any code they have created during the day, for example from practice exercises, to their own GitHub account. Additionally, we made sure this aspect is covered in a dedicated session to look at the challenges of reproducibility from the re-users perspective. Diarmuid prepared an excellent practical exercise for Reproduce This!, where participants had to recreate the following graph based on API Police Data (Stop and Search statistics by Age and Gender).