It is in the Diamond paper that I referenced before: “It is certainly clear from the data reviewed, and the new data and analysis presented, that a massive increase in available pornography in Japan, the United States and elsewhere has been correlated with a dramatic decrease in sexual crimes and most so among youngsters as perpetrators or victims.”
n=4 (countries) is not enough to draw any robust conclusions. That not even enough to run a linear regression. Even conclusions drawn through linear regressions don’t replicate well.
Counting the reported amount of sexual abuse is problematic. It can a sign that people are less likely to report crimes that is in the case of the data for Japan particularly concerning as he suggests: “in these latter years the rapist was less likely to be known to the victim; proving lack of consent became easier.”
The paper doesn’t look like a regular academic paper. It has no abstract. The journal in which is published is named: “Porn 101: Eroticism, Pornography, and the First Amendment”. 101 isn’t a usual name for a journal. The fact that first amendment comes up in a journal name suggest that the journal is politically motivated. If I google the journal name + “imprint factor” I get no results.
Even if you would grant that increased pornography as such doesn’t increase child abuse by pedophiles, it might be still better to have the pedophiles being exposed to adult porn than child porn.
If you’re looking for a regular academic paper, this paper publshed in the Archives of Sexual Behavior appears to make the same point Josh was making. And that journal does appear to be a legitimate peer-reviewed academic journal.
Usually, the more relaxed law on pornography is a result of general sexual liberation of the society and, consequently, there is also higher avaibality of real “non video” adult partners. So, thoretically, the porn could intensify lust, but bigger pool of available real-life adult partners can counter the effect. Some abusers, which are not really pedophiles, but use children as substitute object, are also removed from the game by availability of real-life adult partners.
However, what is the situation with child porn specifically ? Is it a stimulant or inhibitor of the crime ? Are there any countries, which first had availability of adult pornography only, and later lifted the ban on child pornography as well ? What were the statistics of child abuse before and after ?
All I am trying to say is, that the 2 papers from Diamond are not a proof for me, that the availability of child porn leads to lower rates of real-life child abuse. (Intuitively, I guess it is probably so, but it is not proven.)
n=4 (countries) is not enough to draw any robust conclusions.
That’s pretty good for studies where we are counting “nations” to come up with our N.
Counting the reported amount of sexual abuse is problematic. It can a sign that people are less likely to report crimes that is in the case of the data for Japan particularly concerning as he suggests: “in these latter years the rapist was less likely to be known to the victim; proving lack of consent became easier.”
He is certainly aware of the issue. I think the passage you quote strengthens rather than weakens his conclusion in that case.
The paper doesn’t look like a regular academic paper. It has no abstract. The journal in which is published is named: “Porn 101: Eroticism, Pornography, and the First Amendment”. 101 isn’t a usual name for a journal.
Right, it’s a book, not a journal. When access to journal articles requires payment, citing them is problematic.
The fact that first amendment comes up in a journal name suggest that the journal is politically motivated.
There may be some bias in the book. Social science research in general is very politicized, and sex research more than most. Since these findings have potential implications that run counter to received wisdom on child pornography, the most eminent researchers who don’t want to lose their grants might be reluctant to do this sort of work. All sex research has to be examined keeping in mind the political goals of the authors, including all the work on the harm done by pornography.
That’s pretty good for studies where we are counting “nations” to come up with our N.
Putting in effort in no way implies that you end up with the truth. If you want to know the truth you have to look into the underlying statistics. The underlying statistics don’t care that it’s hard to get data about multiple countries.
There no reason to look at countries. Crime statistics are available for US states. You have 50. Maybe you can also find data about pornography sales for each of those states.
There’s Google Trend data that you could use to find out how pornography distribution differs between US states. Google Trend data might even tell you something about the amount of child pornography in relation to other pornography.
You could add some sort of crime like theft to control for difference in the crime rate that aren’t sex related.
You could also control against factors that people frequently use to explain changes in amount of sexual assault. I’m sure the literature on that topic will suggest a few ideas that you should control for.
When access to journal articles requires payment, citing them is problematic.
Usually journals have freely available abstracts of their articles. There are also resources such as http://www.reddit.com/r/Scholar that provide access to articles for everyone.
Reading acadmic papers is a good way to increase one’s understanding of how the world works, even if they aren’t always perfect.
I believe that they did look at crimes like murder and assault as a control for sex crimes in at least some cases.
I did hear of a study once (no, I don’t have a citation) tracking US sex crime rates in relation to when the internet (broadband?) became widely available in different parts of the country, finding some tendency for rape to go down after the internet was available.
In any case, those are all helpful ideas for professional sex researchers but go beyond my competence.
n=4 (countries) is not enough to draw any robust conclusions. That not even enough to run a linear regression. Even conclusions drawn through linear regressions don’t replicate well.
Counting the reported amount of sexual abuse is problematic. It can a sign that people are less likely to report crimes that is in the case of the data for Japan particularly concerning as he suggests: “in these latter years the rapist was less likely to be known to the victim; proving lack of consent became easier.”
The paper doesn’t look like a regular academic paper. It has no abstract. The journal in which is published is named: “Porn 101: Eroticism, Pornography, and the First Amendment”. 101 isn’t a usual name for a journal. The fact that first amendment comes up in a journal name suggest that the journal is politically motivated. If I google the journal name + “imprint factor” I get no results.
Even if you would grant that increased pornography as such doesn’t increase child abuse by pedophiles, it might be still better to have the pedophiles being exposed to adult porn than child porn.
A friendly nitpick: I think you meant “impact factor”. That doesn’t yield results either, of course.
If you’re looking for a regular academic paper, this paper publshed in the Archives of Sexual Behavior appears to make the same point Josh was making. And that journal does appear to be a legitimate peer-reviewed academic journal.
Usually, the more relaxed law on pornography is a result of general sexual liberation of the society and, consequently, there is also higher avaibality of real “non video” adult partners. So, thoretically, the porn could intensify lust, but bigger pool of available real-life adult partners can counter the effect. Some abusers, which are not really pedophiles, but use children as substitute object, are also removed from the game by availability of real-life adult partners.
However, what is the situation with child porn specifically ? Is it a stimulant or inhibitor of the crime ? Are there any countries, which first had availability of adult pornography only, and later lifted the ban on child pornography as well ? What were the statistics of child abuse before and after ?
All I am trying to say is, that the 2 papers from Diamond are not a proof for me, that the availability of child porn leads to lower rates of real-life child abuse. (Intuitively, I guess it is probably so, but it is not proven.)
That’s pretty good for studies where we are counting “nations” to come up with our N.
He is certainly aware of the issue. I think the passage you quote strengthens rather than weakens his conclusion in that case.
Right, it’s a book, not a journal. When access to journal articles requires payment, citing them is problematic.
There may be some bias in the book. Social science research in general is very politicized, and sex research more than most. Since these findings have potential implications that run counter to received wisdom on child pornography, the most eminent researchers who don’t want to lose their grants might be reluctant to do this sort of work. All sex research has to be examined keeping in mind the political goals of the authors, including all the work on the harm done by pornography.
Putting in effort in no way implies that you end up with the truth. If you want to know the truth you have to look into the underlying statistics. The underlying statistics don’t care that it’s hard to get data about multiple countries.
There no reason to look at countries. Crime statistics are available for US states. You have 50. Maybe you can also find data about pornography sales for each of those states. There’s Google Trend data that you could use to find out how pornography distribution differs between US states. Google Trend data might even tell you something about the amount of child pornography in relation to other pornography.
You could add some sort of crime like theft to control for difference in the crime rate that aren’t sex related.
You could also control against factors that people frequently use to explain changes in amount of sexual assault. I’m sure the literature on that topic will suggest a few ideas that you should control for.
Usually journals have freely available abstracts of their articles.
There are also resources such as http://www.reddit.com/r/Scholar that provide access to articles for everyone.
Reading acadmic papers is a good way to increase one’s understanding of how the world works, even if they aren’t always perfect.
I believe that they did look at crimes like murder and assault as a control for sex crimes in at least some cases.
I did hear of a study once (no, I don’t have a citation) tracking US sex crime rates in relation to when the internet (broadband?) became widely available in different parts of the country, finding some tendency for rape to go down after the internet was available.
In any case, those are all helpful ideas for professional sex researchers but go beyond my competence.