Thoughts on Crowdsourced Research data

Here are my reflections from a  short talk on  crowdsourced research data given to PhD students and supervisors last week at Oxford Brookes.

Collecting relevant data for unfunded research projects is getting harder and yet is a requirement for most PhD studies. Getting participants to engage in research projects generally is hard – when I ask my students how many requests they get  to  complete surveys via  apps or online every week  the answer is always ‘hundreds’ followed by but I only do them if they are really interesting or they pay well.

Without resources, both research students and academics  face challenges in getting response data and so some are turning  to online participant platforms to  assist in recruitment and completion rates for their studies, and this might not be a bad thing.

During my talk (available here via slideshare) I outlined some of the ethical challenges posed by platforms such as Amazon’s Mechanical Turk and its  previous heavy use by well established Social Science professors, particularly Americans,  as an approach to gather fast, relatively cheap, and ‘adequate’ quality data.

mturk

Academics publically acknowledging their use of MTurk including Uptal Dholakia – an internationally recognised Consumer Psychology professor who wrote an insightful blog (see here) on his experience as a ‘turker’ (a person who completes online research tasks for a payment) as opposed to one commissioning the task.  Becoming a ‘turker’ changed his views of how he would use Mturk in any future study. Dholakia’s honest blog highlighted  how little ‘turkers’ earn, how long some studies take to complete, how poorly designed some of the studies are which result in poor quality data – including biased questions, and how Mturk is unfairly weighted against those completing tasks.

Whilst Dholakia’s observations are not new and an increasing number of universities internationally and  Marketing journal editors are also questioning the value and ethics of Mturk generated data,  I was impressed by his attempt to venture to the other side as a researcher  and to go public with what he had found.

Leave a Reply

Your email address will not be published. Required fields are marked *