The basis of their survey is depicted here:
https://www.kucoin.com/blog/more-than-half-of-Turkish-adults-invest-in-cryptoWhilst the full public survey resides here (pdf on a Google drive, so I’ll place it in code markers):
https://drive.google.com/file/d/18uoc4-PJmZ36K9ub8UIdacc_OnjS7ajn/view - Google Drive
Disregarding (for now) the full content of the survey, the first aspect to look into is the methodology, being as follows as per their words state in the full report:
This survey report is based on a total sample of 550 adult crypto investors aged 18 to 60 in Turkey. Crypto users are defined as those who currently own or have invested in crypto over the past six months. Survey respondents were polled online from May 5th to May 12th, 2023. The data presented in this report are based on the survey conducted by KuCoin using SurveyMonkey Audience.
Later on, a foot note on a chart also indicates that:
*The incidence rate of crypto investors 52% is estimated based on screening around 1058 Turkish adults and identifying 550 crypto investors among them to complete the survey.
Turkey’s polulation is
close to 86M people. Looking at a recent
breakdown of the population's pyramid structure, I’s say that roughly 60% of the total polulation falls into the age range of the survey (close to 52M people). Playing around with
their sample size calculator ,the sample size used for the survey should suffice (those alleged miracles of statistics), … but …. The key is that the survey panel has to be spared, non-biased and representative.
This latter factor is the key, and I’m not really convinced that it is met. The 1058 Turkish adults were selected using SurveyMonkey Audience (as per their own description). SurveyMonkey allows one to conduct a survey using their panel data (for a cost). The people that conform the panel are generally volunteers that get paid to fill in a survey when contacted to do so (see
https://paidfromsurveys.com/surveymonkey-rewards-review# for example). This creates a first stage of potential bias, as the abstact profile is that of those with a certain technological degree of knowledge, willing to do stuff (i.e. surveys) online. I’d confront that to a random set of 1.058 people chosen at an average market. The latter should prove to be more heterogeneous in profile and knowledge (or lack off).
To contrast,
another survey stated that figure in 2022 was of around 5,5%, whilst
Statista set the figure at 47% for 2023, which backs KuCoin’s results ...
In any case, the figure 52% figure does seem overly high, and it is difficult to assimilate as a fact to be taken for granted without questioning the potential bias of the survey panel.