Content
@
0 reply
0 recast
0 reaction
Daniel Lombraña
@teleyinex.eth
Sometimes the rabbit hole goes deep and you don't know it. Within Warpcast there has been always an argument about what users want vs what they actually do. If you have been here long enough, I'm sure you have read somewhere the "revealed vs stated preferences" discussion about our Warpcast feeds. While this concept has been in the back of my mind, today at work, within a team meeting I remembered it as it fitted quite well within what we were discussing. I had a vague idea about this concept, I searched in Google to see if I could find something related to it so I could learn more. Interestingly there were tons of pages, but I was lucky and found a paper about it https://onlinelibrary.wiley.com/doi/full/10.1002/hec.4246 To my surprise, Wiley has an open access to it and I could send it to my kindle. I didn't check too much as I was planning to read it later today or tomorrow. Well I've finished it & the paper is about how you can use revealed preferences to create better stated preferences surveys
1 reply
0 recast
3 reactions
Daniel Lombraña
@teleyinex.eth
But in which field would you say? In blood donations with UK. The paper tells you how using a large database of donors and their frequencies (revealed preferences) can be used to create a good survey to reduce the bias of stated preferences. What I have liked the most from the paper is that: 1.- most of this work id based on another paper doing something similar but in transportation ( I work at a mobility company so this coincidence is beyond incredible) 2.- how they prototype surveys until finding one that has the less bias by understanding the variables in the survey that introduce the bias ( which is f**king amazing because they're using a lean approach to surveying). 3.- that most of us do surveys without thinking carefully about how well done they should be to represent something properly. If you're interested in understanding more about this, check the paper. You won't regret it.
1 reply
0 recast
1 reaction
