He would not have made that deceson at alpha=0.1 probably, since that would correlate to a 10% chance that the results he saw were caused by random smapling variation with the same null hypothesis. When lives are on the line, I’d argue even 0.05 is too high. I’d strive for anything below 2 or 3 %.
The P value in this context means that there is a 1.1% chance that the results he observed, 89.4% of kids being vacinated, was due to random sampling varition with the null hypothesis of 90% of childred being vaccinated. This is statsitically significant, but not important, since 89.4% is so close to 90% that we really do not care about the slight variation.
type 1
type 2
high alpha level, since its a lower cutoff for users, and therefore would seult in the bank rejecting the null less often
type 1 error up, 2 down
The banks ability to reject someone who would not pay back their loan
raise cutoff score
increase other type of error