This following link contains a more in-depth (and statistical) description of the experiment and its results at the time they were measured, bettering the content found on
this other document, and written by the authors of the experiment themselves:
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2822729Just to highlight a couple of things:
- The main goal of the experiment was to see what people did with their airdropped bitcoins after 2 weeks.
- For contrasting purposes, they created to groups: the first group received their BTCs on day x, whilst the second group received it two weeks later.
- They tracked cashing-out through the digital wallet provider (for those that did not resort to a custodial wallet – 29% used an open-source solution – presumably custodial).
- They added other variables to study the groups from different perspectives (social network, attitude towards bitcoin, demographics, whether they were new to bitcoin (89% were) and so forth.
Essentially, nearly 10% cashed-out within two weeks after receiving their bitcoins, where the segment of so called early-adopters (see below) did so at higher rates (10,8% for those in the first group, 18,3% for those in the second).
There are some specifics I personally don’t like in the study. For example, the use of the label "early adopter" is given to those that signed-up earliest to the experiment (the first 25%). The assigned half of these early adopters to each of the two main groups (as said, group 2 received there bitcoins two weeks after group 1). Considering they had 5 days to sign-up for the experiment, it seems to me like an overkill tag. They are early birds, but not early adopters.
Additionally, the document talks about studying how students rejected (or accepted bitcoin), associating
rejection to selling bitcoin for USD. That is by far a wrong concept, or at least, choice of words, even likely so back in 2014 when the experiment took place.
I haven’t taken a look at the data in detail, but the second group of early adopters likely cashed-out significantly more in numbers due to the influence of those in the first group and/or variation of price in between cashing-out windows of the two groups (I haven’t contrasted this hypothesis though).