Expected value utility is of course the expected value of income p*X+( 1-P)*Y, and the utility of this value is of course called expected value utility.
Their comparison refers to whether people usually choose between expected return and risk return. When the expected utility is greater than the expected utility, they can understand his decision through repeated choices. If he can choose n times, according to the law of large numbers, the utility is approximately equal to N*[p*U(X)+( 1-P)*U(Y)], which is greater than the n * u that keeps the expected utility.
The utility of the expected value directly puts the expected value into the utility function, which means that this person regards the expected value as a "definite" value, and after processing, he gets a definite utility (note that what he gets is an realized value of the utility function, which is of course certain).
Expected utility is the expected value of utility, and the expected value is the distribution mean with variance, so the expected value of utility is a risky situation.
Its practical significance is: expected utility, that is, what I think utility is, is a subjective concept; Expected value utility refers to a relatively objective value that can be obtained according to certain scientific and reasonable methods, which represents the actual utility of goods. Then if I think my expected utility or my satisfaction is less than the expected utility, I think it is not cost-effective, which shows that I am avoiding risks.