- This event has passed.
Methodologies for Global Development Research,
Event Navigation
A Global Development Research Division workshop
Ethnographic perspectives on the drug trade – Tom Grisaffi
How can you collect data on illegal activities? Tom will give a brief history of ethnographic methods and how he has used this approach to study politics and cocaine production in Bolivia and Peru. He will highlight three key areas: how to negotiate access, how to conduct fieldwork, and how to write up the data.
The ‘difference in differences’ methodology for impact evaluation – Stefania Lovo
‘Difference in differences’ is a quantitative method for impact evaluation that uses observational data from before and after a particular event took place. It is widely used because it is intuitive and fairly flexible. In her presentation, Stefania will provide a non-technical overview of how the method works and its main advantages using examples from her own research.
The walking interview: experiences from Mozambique – Alex Arnall
In recent years there has been growing interest in the walking interview method, during which the researcher accompanies people as they move around the landscape. Alex will share experiences of carrying out walking interviews with small-scale farmers in Mozambique – what the strengths of the method are plus some of the potential pitfalls.
When no does not mean no: a story of gender preferences in online vacancies – Tho Pham
Tho will highlight the applications of machine learning in economics/development research with reference to a work-in-progress where she looks at gender preference hidden in online job ads. First, she will describe the data collection procedure where job vacancies posted on various job sites are scrapped on daily basis. Second, she will briefly discuss some machine learning techniques, namely semi-supervised Latent Dirichlet Allocation (LDA) and Random Forest (RF), which are used to examine the job ads’ text to determine embedded gender preference.