DTC Workshop: Analysing Data with Crowds
Tuesday 25 November 2014, 09:45:00 - 16:00:00
Seminar Room, Oxford Internet Institute, 1 St Giles’
Register by emailing: email@example.com. Maximum: 20 students. This course is open to all DPhil students and is offered in conjunction with the DTC Workshop: Capturing Online Data on the Monday. Attendees can choose to attend one or both days.
This course is designed to give students hands-on experience with the power and the pitfalls of using crowd sourced labour in social science research. Crowd sourced labour is becoming increasingly attractive as a part of the researcher’s toolkit.
This course is designed to give students hands-on experience with the power and the pitfalls of using crowd sourced labour in social science research. Crowd sourced labour is becoming increasingly attractive as a part of the researcher’s toolkit. When data can be broken into small chunks and tasks can be clearly specified for the layperson, crowdsourcing can be a power and inexpensive way to scale up the coding of data. In general, people are paid for “micro tasks” that are small, discrete tasks that add up to big numbers. Yet if microwork is not performed correctly, the data can be of poor quality or downright unusable.
This one day session will demonstrate the power of crowdsourcing by showing how to take data and prepare it for microwork in Excel, how to choose the appropriate crowdsourcing platform, how to bid for tasks using our chosen example platform, how to lay out a page and how to sketch out the appropriate scale for tasks. We will also reflect on who the “crowd” behind the platform really are, and what implications this has for research design and ethics.
Students will get hands-on experience with performing crowdsourced coding on a commercial platform at no cost to the student. Having such real world experience will be extremely useful for students considering whether to use crowdsourcing in their own work and at what cost.
Students taking this one-day course may wish to consider also taking our one-day course on “Capturing online data”, in which they learn how to acquire large numbers of social media messages that can then be coded using the methods taught in this course.
|Part 1||Introducing crowdsourcing (30 mins)|
|Part 2||Designing a crowd sourced project (1 hour)|
|Part 3||Preparing and launching data (1 hour)|
|[Lunch break]|| |
|Part 4||Monitoring data and considering data quality. (30 mins)|
|Part 5||Reviewing completed data. (1 hour)|
|Part 6||Reflections on crowd sourced data / class discussion. (1 hour)|
**Attendees should note that they will need to bring their own laptop to the session. If you aren’t able to supply your own laptop, please email firstname.lastname@example.org as there are a limited number available to borrow within the department.