I'm still not following the logic here.

Your study cites that it collected data from 1,005 participants. You mention that 'Shudder' and 'iflix' were cancelled by 0.23% of participants, respectively.

0.23%/1,005 = 2.3115... people. I didn't know that a fraction of a person could fill out a survey?

​

Your numbers above also all sum to 100.00%. To be clear, is this to say that each respondent, other than the ones that selected 'None of the above', all claim to have cancelled only one service, and not a single one has cancelled more than one service? (i.e. of those with video streaming services, 33.08% cancelled no services, 66.92% cancelled one service, and 0.00% cancelled two or more services?)

These figures don't make sense. according to this survey, about 67% of people canceled in the last 12 months... Assuming that there is zero overlap in subscribership among three of the most popular services (very unlikely), these only account for about 26% of the services cancelled. What totals up to +/- 40% of the other streaming services that were cancelled? (Which btw, would also need to have zero overlap with one another)

Also, I have a hard time believing that amazon lost 9% of its subscribers in the past 12 months. According to Statista, Amazon continued to gain household subscribers in the US, each and every year for the past several years.

https://www.statista.com/statistics/861060/total-number-of-households-amazon-prime-subscription-usa/