I tried to further understand the report by relating one of the talking points to the underlying data. There are other taking points in the report, that are all very important, by looking at this one I am not commenting on any of the others.
95% of men who answered yes to the question “Have you had sex or sexual contact with a person below the age of 18 while over the age of 18?” were over 24 years of age, and almost half (48.4%) were over 54 years of age.
On the face of it, this statement reads as if people over 54 years old are 10x more likely than those between 18 and 25 to have sexual contact with people under 18. The younger age group has a 5% number when the older group is nearly 50% - 10x bigger
But the author is comparing the sum of all the people in a 6 year age range with all the people in a much wider age range of 55-64 and 65+. No upper limit is given to age - it could be 75 or 80 years. The comparison is between a band that is 6 years wide and a band that could be 25 years wide.
The report does give a breakdown of the age range of the survey participants (figure 9). It is not fully clear but it looks like they then take a set of Australian census data and use this to weight the original survey so that it reflects the make up of age ranges in the census (supplementary table S1).
The tables show that when scaled the 18-24 age group represents 15.2% of the population and the 55+ group is 34.5%. If target subjects were evenly distributed across age you would expect these percentages to be reflected in the answer to the 'below age of 18' question asked in the survey.
So, 18-24 expect 15.2%, survey reports 5%
54 and above, expect 34.5%, survey reports 48.4%
The numbers for 54+ age group are higher than expected but only by 40% or so. This difference may just be due to a change in attitudes - some of these people could be 75 or 80 years old - they turned 18 in around 1960. Attitudes have changed a lot since 1960
The low value for the 18-24 age range could also reflect changing attitudes.
Looking through the report, the original survey contained 275 people in the 18-24 age range. Of these, around 5% are relevant so the group size is about 13. It would not take much underreporting to significantly skew this group
Reasons for underreporting? Maybe the people in the 18-24 age range are the ones most likely to still have a partner who was 18 or under when they met - perhaps these people underreport because the survey questions imply their relationship is one of sexual predation?