Quote:
Originally Posted by Vic 2.0
This is the best I could find at the moment. If it's not enough, I guess we'll have to agree to disagree.
|
Somehow it never occured to me to check if the research their book was quoting was their OWN research. I probably would have found the Meta Analysis myself had I thought to check that. I just assumed if they were citing research in a book then they were citing someone elses.
First issue here.
The Analysis however suffers from one of the most basic errors I can think of. It does not mention the selection criteria for how the studies included in the Meta were picked. Found yes. Selected no.
If for example I have 100 studies and I do a meta analysis of 40 of them - I should pick my studies randomly. What I should not do - for example - is find the 40 that most support the claim I want to make - and only include them in the Meta. THAT is why selection criteria need to be mentioned in any Meta. They are not mentioned here. That is NOT good.
For example in the document they say openly that studies that do not support their postition were left out....
Quote:
Though studies from Denmark tend to be cited to support the contention that availability of pornography is related to lower rates of sexual offending, those located by the researchers did not meet the criteria for inclusion for the present meta-analysis.
|
.... but convieniently we are not told what that criteria was. Was it "papers that agree with US" for example?
Second issue here.
Worse I randomly looked at some of the studies they did include. These are not great. Take for example:
"Violent pornography and self-reported likelihood of sexual aggression. Journal of Research in Personality".
The first one basically asks people who have looked at porn if they personally feel any subjective increase in a tendency towards thinking about rape. So the hell what? THINKING about it is not the same as an increased chance of engaging in it.
Further they speak not just of rape but "Twenty-seven percent of subjects indicated some hypothetical likelihood of raping or using sexual force against a woman". Thats pretty vague is it not? What about consensual violence such as in S&M? Saying you would like to engage in a violent but consensual sexual encounter is not a bad thing but it sounds like such is included here.
Finally even if a correlation is established between those who engage in violence in sex and those who watch violent porn - how does this establish the latter led to the former? Perhaps it is an interest in the former led them to the latter?
Third issue here.
"
The effect of polygraphy on the self report of adolescent sex offenders: Implications for risk assessment. Annals of Sex Research" is worse again as this is not even a study of the effects of porn on agression. It is a study of the effects of using polygraphy on people who were already sexual perpetrators!!!
Hardly fair to base an assesment of pornography as a whole on the basis of a report only looking at sexually deviant covincted sex offenders is it
?!?!?!?
Further this stufy makes the same correlation-causation error where they found that there is a correlation between the number of victims and the kind of porn used. Again it is just assumed that the latter led to the former and that the former has nothing to do with the latter. Perhaps whatever gave them the propensity to offend more also gives them a propensity towards violent pornography.
Summary
The methodology of this paper is suspect, especially how they chose the studies to include. As are the papers within it - such as including a study about polygraphy, not pornography, which studies ONLY sex offenders and not your common joe on the street - or any such control.
At best the Meta establishes a weak and tenuous correlation between VIOLENT pornography and violent acts. There is nothing addressing whether one is the cause of the other or share root causes however. Nor does establishing a correlation to VIOLENT pronography support your contention that pornography AS A WHOLE is dangerous.