If no physical force has been used, the question remains open if it was a volitional relation or adult power has been misused. In this situation it is reasonable to assume that the probability of power misuse is higher if the power difference is higher.
Let's note here an essential difference: this assumption does not mean that an incest relation is worse. There may be a volitional incest relation and power misuse in a non-incestuous relation. We are talking here only about probabilities. Thus, our hypothesis is that
Higher power difference makes volitional relations less probable.
Now, there are three main aspects in adult-child-relations which may increase the power difference: incest, gender difference and the age of the child. Thus, our hypothesis suggests that these differences lead to higher probability of unwanted sex, thus, higher negative impact.
From this point of view, we have one difference (adult-child), for man/boy, two (adult-child, male/female) for man/girl, and three (adult-child, male/female, parent/child) for father/daughter incest. The hypothesis is that the probability of volitional relations is lower for girls than for boys and lower in cases of incest. This prediction is in agreement with the meta-analysis of Rind et.al. 1998. They note a clear gender difference: boys react much more positive. They have also found that incest is the second abuse-related factor after the use of force which influences the effect size.