I'm writing a master thesis in the subject of technology trajectories affecting the variance in cryptocurrencies and need to have reliable data.
Question is if there is any consensus about which one is more trustworthy than the other?
Strange question. A consensus is a collective perception, it's most unlikely you'll get any
reliable data by taking this approach.
I predict you will need to define carefully your concept of reliable. There is a key difference between alts and stocks in that (AIUI) maintaining and submitting a full record of each and every stock trade is obligatory and mandated by legislation. No such accounting reliability attaches to altcoin trading. Trading records must be assumed to have an indeterminate degree of inaccuracy and to be, overall, incomplete.
The term market capitalization is only
analogous in this context and in order for the analogy to be perceived as useful, the mapping has to be fudged: equal to the share price times the number of shares outstanding. The argument shares === coins has not been accepted unconditionally by everyone, I'm given to understand that establishing a market cap for Ripple remains a contentious issue.
I recommend you peer into the horse's mouth so to speak and start inspecting the contents of the data streams emitted by the various exchange APIs (where an API exists, else you'll need to pagescrape). That will give you the most reliable data.
Cheers
Graham