We'd need to dig through the study itself and whatever information was available about it. But typically what you'll find what was that responders to the survey

We'd need to dig through the study itself and whatever information was available about it. But typically what you'll find what was that responders to the survey answered a series of questions about storage & anxiety. The researchers then concluded 37% of their total anxiety came from the data stored. How this was reached is missing from this report it seems. Also there should be some sort of uncertainty or confidence interval associated with it.