Crowdsourcing is increasingly used in science as a means of obtaining, transforming, and analysing data. We ask what we can achieve with crowdsourcing that they couldn’t achieve before?

Crowdsourcing, defined as outsourcing a task to a crowd through the Internet, is increasingly used in science and engineering as a means of obtaining, transforming, and analysing data. In this half-day workshop we ask, what can social scientists achieve with crowdsourcing that they couldn’t achieve before?

Using a combination of lectures and practical walk-throughs, we will examine the data collection applications of microtask-style crowdsourcing, such as social surveys, experiments, human search, and human sensing, as well as its applications in data analysis and transformation, such as coding, categorization, image analysis, transcription, and translation. We will discuss both the benefits and the limitations of crowdsourcing – what it should and should not be used for in social research – and highlight key practical challenges such as quality control. We will also peek behind the curtain to look at the crowd workers themselves, and discuss the ethical issues involved.

Participants’ main takeaways are:

  • An understanding of the new opportunities offered by crowdsourcing for social scientists, as well as its main limitations and ethical considerations

  • A basic practical idea of how to use one or two of the most common commercial crowdsourcing platforms for data collection and analysis tasks likely to arise in social research

The workshop is structured in the following way:

Table: Workshop programme

Time

Title

Description

09.30 – 10.15

Crowdsourcing and microwork: an introduction for social scientists

This lecture-based session will introduce the phenomenon of crowdsourcing, provide some useful definitions and distinctions, and place these buzzwords in the context of some older academic discussions on participation. The session will also showcase some of the most prominent uses of crowdsourcing in business and academia.

10.15 – 11.30

Collecting data through crowdsourcing

In this combination of a lecture and walk-through, participants will first receive an overview of the different types of data collection approaches feasible with crowdsourcing, and then follow and participate in a practical example of one such approach using a commercial crowdsourcing platform. Based on the data collected and on findings from the presenter’s earlier research, participants will learn about who some of the crowd workers actually are and how crowd work fits into their lives.

11.30 – 11.45

Break

11.45 – 13.00

Analysing data through crowdsourcing

In this session structured similarly to the previous one, participants will learn about different data analysis approaches that make use of crowdsourcing. They will then follow and participate in a practical example of using a commercial crowdsourcing platform to analyse the data collected during the previous session. Participants will also discuss the potentials and limitations of the data analysis approaches introduced in the session.

About the speakers

This page was last modified on 15 March 2017