|dc.description.abstract||I conducted an exploratory research of the little understood principal-agent interaction
in a reward-based crowdfunding environment. A broad-stroke exploratory research like
this is unavoidably limited in the extent to which any of the findings can be generalised
beyond individual cases.
I attempt to mitigate this by complementing a netnographic approach with a sentiment
analysis classifier programme that I have developed. This holistic approach allowed
me to gain deep insight into the mechanisms that allow the principal-agent problem on
Kickstarter to be resolved successfully despite lack of rigid legislative regulation.
I find that backers on Kickstarter possess sufficient tools to minimise information
asymmetry and thus, the principal-agent problem.
I further discover that formations of backers are comprised of two distinct groups – a
small vocal and a much larger silent one, with the latter adjusting to the opinions of the
former in the short term. I also find that the crowdfunding platform plays a limited, yet
important role in resolving the principal-agent tensions.
Finally, I find evidence that formations of backers that surround Kickstarter projects
are fluid in their structure, exhibiting under different conditions features of both communities