Note that for any real number \( x \in (0, 1) \), and \( n > m \)
we have
\( x^n < x^m \)
Therefore we have
\( X^n < X^m \) for any random variables with support \( (0, 1) \)
and thus
\( E[X^n] < E[X^m] \)
In particular,
\( E[X] > E[X^2] > E[X^2] - E[X]^2 = Var[X] \)