This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.
Key Information
Please note this course is not available in the 2022-23 academic year
About
Summary
This option puts forward the experimental method as an essential element of social data science, which offers the potential to explain patterns or irregularities in human behaviour. First, large-scale data may reveal natural experiments where the effect of a change in platform design or commercial offering can be observed in the data ‘as if’ it had been generated randomly (Dunning, 2102). Second, explicit interventions may be made in Randomized Controlled Trials (RCTs), using the internet or mobile platforms as a ‘field’ to explore the potential effect of commercial or policy interventions. Experiments of both kinds may be used to understand social influence and network effects on behavioural outcomes.
Key issues to be tackled include:
- What sources of data reveal natural experiments (eg. platform redesign) and what new methods are needed to analyse them (regression discontinuity design, propensity score matching, etc.)?
- How can experiments be run effectively at scale without the cooperation of a platform operator (eg. crowdsourcing survey experiments)?
- Where can/should RCTs be carried out, eg to nudge citizens towards social optimal outcomes, and how can we resolve the ethical dilemmas provoked by experimental methods carried out to scale on digital platforms?
Learning objectives
1. To familiarise students with philosophy of science and theory undergirding the experimental method and the basic elements of experimental design, including an exploration of variations across disciplines in terms of the sort of experiments that are carried out;
2. Understand the logistical and ethical challenges of the experimental method, and in particular to understand strengths and weaknesses of the state-of- the-art scientific approaches used in the social science experiments focused on online settings;
3. Develop student competence in both identifying natural experiments, and in designing novel and viable online experiments.
Learning Outcomes
This module will prepare students to interpret and practice the experimental method, an increasingly important area of social science research in general, and in social data science in particular. It will also illustrate to them how experimentation may be incorporated into policy design – including where experiments may be a ‘natural’ by- product of policy change – thereby enriching any policy research or analysis they carry out in their future working lives.