It seems premature to assert such precise percentages on identity document verification habits based solely on 78 responses.
A survey of this size, although relevant to initiate reflection, doesn't provide a robust enough basis for statistical generalizations about the population.
It would be prudent to consider a larger and more representative study to validate such claims, especially if we want to understand the stress factors related to these procedures.
For example, a national survey of 1000 people would provide much better reliability for these percentages.