Gail Dines: Short on Facts, Crap Figures

*** Warning – Very long post ahead. Get supplies before sitting down to read ***

Last time around I followed the trail of just one dubiously provenanced Internet porn ‘statistic’ from its first appearance in 2003 on a US website owned by tech entrepreneur and content filter salesman, Jerry Ropelato, through its subsequent appearance in everything from academic literature to an official report published by the International Telecommunications Union, all without anyone, it seems, bothering to check the validity of that ‘statistic’ for which there is currently no known original source.

I put that last article together to serve as a precursor to this one, which looks in detail at a variety of statistics and other items of ‘factual’ information published on a ‘Facts and Figures‘ page on the website of anti-porn activist Gail Dines’ “Stop Porn Culture” website.

There is a bit of background to this of course.

Within the last couple of weeks or so the academic publisher Routledge published the first issue of its new quarterly journal ‘Porn Studies‘ – and if you do wander over and take a look, all the articles in this first can, at least for the time being, be downloaded free of charge.  – and the some of the publicity surrounding the launch contained a reminder of something Dines said last year, when it was originally announced that the journal would be going into print this year:

Gail Dines, an anti-pornography activist and professor of sociology and Women’s Studies at Wheelock College, Boston, told The Observer in June that while it was vital that pornography was studied and research published, she had grave concerns about the editorial direction of the journal.

“These editors come from a pro-porn background where they deny the tons and tons of research that has been done into the negative effects of porn,” she said.

“They are akin to climate-change deniers. They’re taking a bit of junk science and leaping to all sorts of unfounded conclusions.”


Gail Dines is accusing other academics of dealing in ‘junk science’ and behaving like climate change deniers?

Well, I think we’ll just have to see about that, and her website’s ‘facts and figures’ page strikes me as being as good a place to start as any for a couple of reasons.

First, as a general point, I’ve always taken the view that the manner in which an academic deals with factual information speaks volumes about their overall standards of scholarship or, to put it more simply, if you can’t get your facts straight then why on earth would you expect me, or anyone else, to trust anything else you have to say.

Second, this particular page is a relatively recent addition to the Stop Porn Culture website.

Tracking changes to the design of the site via the Wayback Machine Internet archive what we find is that the site underwent a complete redesign sometime around February 2013, at which point a ‘facts and figures’ page was added to the site – and I’ve pulled a complete copy of that original page out of the archive and into this PDF document as it appeared in March 2013.

The current version of Dines’ ‘facts and figures page’ made its first appearance somewhere November 2013 – and again I’ve grabbed a PDF copy of the page for posterity (and as a hedge against any possibility of stealth editing) – so irrespective of the apparent age of any of the material included in it according to the citations and references on the provides what we have here can reasonably be taken as a fair representation of what Dines regards as factual information based on her own personal judgement.

What follows is, therefore, an evaluation of the veracity and reliability of that judgement.

So let’s get started with our first ‘fact’:

There are over 68 million daily searches for pornography in the United States. Thats 25% of all daily searches (IFR, 2006).

Right from the off we’re into one of Jerry Ropelato’s unsourced figures that dates back to at least 2003, although somewhat curiously Dines claims that this ‘statistic’ relates only to searches for porn in the United States, which is not and never has been something that Ropelato has claimed on his own. So this particular ‘statistics’ would appear to have been amended by Dines, perhaps in an effort to make it appear a little less implausible than it actually is – and trust me it really is implausible.

This is also – if you’ve read my previous article – the ‘statistic’ that found its way into the 2010 Papadopoulos review of the Sexualisation of Young People, which was commissioned by the Home Office.

This is a claim that is easily blown out of the water, starting with this graph which shows the estimated growth in global daily search traffic since 1998.


This one’s easy enough to follow as you’ve only got three lines to get your heads around.

The blue line shows the growth in Google’s estimated global daily search traffic since 1998 and it’s based on data I’ve personally pulled together from Google, Comscore Media Matrix and several other web metrics companies. The orange lines shows an estimate of the total global daily search traffic based on the data for Google and estimates of the company global market share over time, and where possible I’ve relied most heavily on figures supplied by Comscore Media Matrix as it’s one of the few companies that appears to have a reasonable handle on search traffic levels in the rapidly growing Russian and Chinese markets. As you can see from the two trends, Google’s overall share of the global market has decline somewhat in the last 3 years and much of that is due to the rapid growth in search traffic in Russia and China where the search market is dominated by local search engines Yandex (Russia) and Baidu (China).

As for the black dotted line, that shows how the Ropelato ‘statistic’ relates to both trends. The claim there is that there are 68 million porn-related searches a day and that these make up a quarter of all search traffic, giving an estimate for the total amount of daily search traffic of 272 million searches. As should be entirely clear from the graph Google, alone, surpassed that amount of daily search traffic as long ago as the beginning of 2004 while the estimate for the total size of global search suggest that even if there is an original source for that claim then it can date to no later than the beginning of 2001 and may well date all the way back to the turn of the millennium.

So, at the very least we can say with a considerable degree of certainty that this particular claim is so far past its sell-by-date that you might just as well be trying to evaluate the form of racehorses ahead of this year’s Derby by looking at results from Rome’s Circus Maximus for all that it has any relevance to levels of search traffic in 2014, but even so we still have the historical claim that maybe twelve or thirteen years ago porn generated around a quarter of all daily search traffic to tackle.

Well, for that one we need to turn to this 2006 study of Search Engine trends which pulls together data from nine individual studies spanning five different search engines over a period of eight years from 1997 to 2005, and to keep thing simple I’ve plotted the trend data for two areas of activity; searches linked to commercial activity and, of course, searches for porn and other sexual material:


As the graph shows, the overall trend in commercial search activity is somewhat erratic, which is likely to reflect both difference in the user base of different search engines over time and, to a degree, the boom-bust-boom passage of the Dot Com bubble. However when it comes to searching for porn there is a very clear downward trend over time, from 16.8% of Excite’s search traffic in May 1997 down to just 3.8% of Dogpile’s search traffic in May 2005.

What’s happening here is not, of course, a decline in the overall demand for online porn but a significant increase in the number of people using the Internet, and search engines, to pursue an increasing range of non-sexual interests, but what this graph does indicate is that if there ever was a point at which fully a quarter of all search traffic was porn-related then this would certainly have been no later than early 1997; and in fact if you were to project the trend line backwards to the point at which it hits 25% then you’ll find that you’re probably looking at a date somewhere around mid-late 1995.

Item 2 on Dines’ list claims that:

The sex industry is largest and most profitable industry in the world. “It includes street prostitution, brothels, ‘massage parlors’, strip clubs, human trafficking for sexual purposes, phone sex, child and adult pornography, mail order brides and sex tourism – just to mention a few of the most common examples.” (Andersson et al, 2013)

And here we have direct link to a source further down the page albeit that the source in question is to be found the website of Melissa Farley whose reputation is, if anything, even worse than that of Dines – both share the unusual distinction of having large chunk of their ‘expert’ testimony in court cases rejected by the judge on the grounds that they were too obviously biased to be credible expert witnesses.

That said Dines’ source in this case is not a piece of research but a pamphlet produced by a Swedish radical feminist group and despite the fact that she’s citing an obviously sympathetic source she still can’t resist trying to gild the lily by tweaking what the pamphlet actually says, which is:

Today sex trade is one of the largest and most profitable industries in the world. It includes street prostitution, brothels, ‘massage parlors’, strip clubs, human trafficking for sexual purposes, phone sex, child and adult pornography, mail order brides and sex tourism – just to mention a few of the most common examples.

Clearly, just being one of the largest and most profitable industries in the world is not enough to satisfy Dines’ purposes, so with a little judicious cherry picking it become the largest, even though that’s not what her source actually says and, in any case, even that claim is made without providing any kind supporting..

So how big is the sex trade?

The simple fact is, no one knows, least of all Dines or the Swedish rad-fem collective that produced that pamphlet because there are no accurate or reliable figures on the scale of the global sex trade for anyone to work with.

The only thing we can actually take from this particular claim, as it appears on Dines’ website, is that the manner in which it’s been reproduced from its original source speaks volumes about Dines’ overall approach to dealing with ‘factual’ information.


In 2010, 13% of global web searches were for sexual content. This does not include P2P downloads and torrents. (Ogas & Gaddam)

The citation here is to Ogas and Gaddam’s thoroughly enjoyable book ‘A Billion Wicked Thoughts‘, from which this paragraph describes how they obtained their data.

We collected about 400 million different searches that were entered into the Dogpile search engine from July 2009 to July 2010. We collected these searches through a process called scraping: we wrote a program to capture the searches listed on SearchSpy, a Dogpile-run Web site that displays in real time the actual searches people entered into the Dogpile search engine. If you visit SearchSpy, it’s like looking through a window into a planetary stream of human consciousness—and you won’t have to wait more than a few seconds to see its sexual side. Of the 400 million searches we collected, about 55 million (roughly 13 percent) were searches for some kind of erotic content. These sexual searches represent the desires of roughly 2 million people. Two-thirds are from the United States, though some users are from India, Nigeria, Canada, and the United Kingdom.

400 million searches collected over 12 months might sound like a hell of a lot of data but in practice it’s the equivalent of around one tenth of the total traffic that was passing through search engines every day back in 2010, so the leap from Ogas and Gaddam’s figures to making a claim about global traffic for a whole year is stretching credibility. It’s a good sized sample but at the scale of traffic we’re looking at over a twelve month period there is still a considerable margin for error in any statistical information you care to extract from it.

You’ll also notice that Ogas and Gaddam refer to ‘erotic content’ and not just porn, a point that’s easily illustrated using this Google Trends for 2012 which compares the relatively levels of search traffic generated by two primarily male brands – Playboy and Maxim – with two where the audience is dominated by women; the erotic novel “50 Shades of Grey” and the lingerie company “Victoria’s Secret”.


And would you just look at that. At its peak “50 Shades” pulled in more search traffic than either Playboy or Maxim, although Playboy does a rather better job of sustaining its popularity over time, as you might well expect for a monthly periodical that’s been running since the 1950s.

As for “Victoria’s Secret” that brand, as you might well expect, generates its peak traffic from mid-November to around the first or second week of December a little ahead of the Christmas holiday period and does so every year.

So, not everyone looking for ‘erotic content’ online is male nor are all such searches looking specifically for porn.

Next statistic…

Pornhub receives over 1.68 million visits per hour. (Pornhub, 2013)

Good for Pornhub but let’s just put that figure into perspective. 1.68 million visits per hour is just over 40 million visits a day or around 1.2 billion visits per month, which seems like a lot until you realise that, according to web metrics company  SimilarWeb, Facebook racks up a little over 23 billion visits a month, while Youtube’s monthly traffic is estimated to be around 15.4 billion visits.

Ah, but hang on a second because there’s a bit of a complication here because the 1.2 billion visits per month that Pornhub are claiming is more than double the number of monthly visits estimated by SimilarWeb (570 million) in its metrics so something is a little off here and from what I can see it’s likely to be the figure given by Pornhub, even if – as we’ll see later – SimilarWeb don’t collect any figures for mobile Internet traffic.

In June 2013, the BBC’s technology correspondent, Mark Ward, contacted what was then called Manwin – it’s since changed its name to MindGeek – the company that owns Pornhub and asked them for their traffic figures and as you’ll see here Ward was quoted a figure of 70 million visitors a day for their entire network of porn sites, which consists of eight porn tubes and a search engine. Using the figures from SimilarWeb for each of nine sites which make up MindGeek’s porn network, four of which are listed amongst the top ten adult websites by SimilarWeb, I came to a daily traffic figure of around 65 million visits, which is in the same ballpark as the figure given to the BBC last year; and from what I can see, MindGeek’s traffic figures look to have been fairly stable over the intervening period. By comparison, when I plugged the much higher traffic figure cited by Pornhub into the same set of calculations and scaled the figures from SimilarWeb for the other eight sites by the same ratio as the difference between the two sets of figures given for Pornhub’s traffic I came out with an estimate for the total daily traffic across the MindGeek porn network of 137 million visits per day, which is way above the figure the company gave to the BBC.

Pornhub also gave data covering it’s UK traffic to The Guardian newspaper in January of this year, although they don’t appear from the article to have specified the exact period that data covers. However, the full results table provides information relating to just over 111 million visits to the site from the UK and, again, if we scale that up to a global figure using Similarweb’s current estimate for the percentage amount of traffic the site gets from the UK (6.7%) we get a figure of around 1.65 billion visits, which is well within the right ballpark for it being based on 3 months traffic at the levels estimated by SimilarWeb of around 570 million visits per month, all of which adds to the impression that SimilarWeb’s figures are rather more accurate that Pornhub’s own claim to be getting 1.68 million visitor an hour.

There is, perhaps, a simple explanation for this.

On the same page that it claims to receiving 1.68 million visits per hour Pornhub also bills itself as the “world’s number 1 porn site” when, in terms of traffic, it’s currently running 3rd in the Adult category in both SimilarWeb’s and Alexa’s rankings, and that does perhaps suggest that a little creative marketing license may have been applied to that figures to better support their claim to be number one.

Now, the next set of claims are particularly interesting because Dines is citing herself as the source:

Globally, teen is the most searched term. A Google Trends analysis indicates that searches for “Teen Porn” have more than tripled between 2005-2013, and teen porn was the fastest-growing genre over this period. Total searches for teen-related porn reached an estimated 500,000 daily in March 2013, far larger than other genres, representing approximately one-third of total daily searches for pornographic web sites. (Dines, 2013).

Okay, so let’s break it down…

Globally, teen is the most searched term.

Bullshit – it’s actually FACEBOOK, and it’s been Facebook since the beginning of 2009 when searches for Facebook overtook the previous number 1 search term, YOUTUBE.


And if you’re wondering why it’s Facebook, the answer seems to be that…

a) a very large number of regular Internet users use Google as their default browser home page , and

b) a lot of those people can’t be arsed to navigate to address bar to type in “” when its easier just to type Facebook in the default Google search box on their home page and then click the link when it comes up.

So, the lesson there is never underestimate the power of sheer laziness.

Anyway, getting back on topic, ‘teen’ clearly is a hell of long way from being the most common search term used by Internet users but what about the next bit?

A Google Trends analysis indicates that searches for “Teen Porn” have more than tripled between 2005-2013, and teen porn was the fastest-growing genre over this period.

Bullshit, again.

Based on a little over ten years worth of Google Trends data, ‘teen porn’ is neither the most searched for or fastest growing porn genre, although you may be a little surprised when you find out what is…


Yes, it’s GAY PORN…

In terms of the overall amount of search traffic going through Google, searches for ‘teen porn’ average out at slightly less than searches for ‘gay porn’ although the difference is small (60-59 in gay porn’s favour) and the upwards trend over time in both searches follows more or less a parallel path. What this tends to suggest – beyond the fact that Kinsey was right when he suggested that a lot more men dabble in a bit of batting for the home team than are ever prepared to admit it – is that the driving force behind the rising trends in searches for ‘teen porn’ and ‘gay porn’ is most likely to be demographic – an increase in the number of people online over time means an increase in the number of people searching for certain very common porn genres just because there are more people online to be searching for porn and not because there’s been any particular growth in overall interest in those genres.

As for ‘teen porn’ being the fastest growing genre, again you’ll see from that same graph that that’s bullshit as well.

Although both started from a much lower base and are much less popular overall, the amount of search traffic looking for both ‘Milf porn’ (5x) and ‘homemade porn’ (7x) has grown faster over the last ten years than either ‘teen porn’ (3.7x) or ‘gay porn’ (2.6x) and there is at least one other porn genre not shown that has seen a sevenfold increase in search traffic over the last ten years, ‘Indian porn’ – there’s the Indian tech revolution at work for you.

And then we come to…

Total searches for teen-related porn reached an estimated 500,000 daily in March 2013, far larger than other genres, representing approximately one-third of total daily searches for pornographic web sites. (Dines, 2013).

And again we have more bullshit although, oddly enough, the raw figure of 500,000 daily searches is pretty much correct – I pulled together the trend data from Google and the traffic figures from SimilarWeb in order to use searches for what is currently the most popular porn site in the world (Xvideos) in order to get a baseline figure from which I could estimate the actually amount of search traffic generated by the term ‘teen porn’ and got an estimate of 536,000 searches per day.

So at least Dines, or whoever she got to do her scut work on this one,  hasn’t completely bollixed the calculation.

That said, even at 500,000 searches a day, the number of daily searches looking for ‘teen porn’ is dwarfed by the number of daily searches for Facebook (113 million), which accounts for around 3% of Google’s daily search traffic compared to just 0.013% which is generated by people that are looking for ‘teen porn’ – and as for the reference point on which these calculations are based, the porn tube site Xvideos, which racks up over 9 million searches per day and a little over 18 times more searches for ‘teen porn’, which does blow rather a big hole in the ‘one-third of all daily searches’ claim.

For the record, the back story here is that Dines’ devotes an entire chapter of her book ‘Pornland’ to developing the risible argument that ‘teen porn’, a term which is clearly understood to mean that the female performers are at least 18 years old, acts as a gateway to child pornography, hence the grossly exaggerated claim that around one-third of all searches for pornographic websites are looks specifically for ‘teen porn’, etc.

And on we go…

The United States is the top producer of pornographic dvds and web material; the second largest is Germany: they each produce in excess of 400 porn films for dvd every week.


Okay, from what I can see this particular ‘statistic’, which is unsourced, appears nowhere else online other than on Dines’ website and one or two Tumblr and Facebook pages that are clearly taking their information from Dines. It doesn’t even appear in Dines’ book ‘Pornland’.

At its peak in around 2005/2006, the US porn industry appears to have been releasing at most around 1,100 new DVD titles a month, which is around 275 a week and not the 400 that Dines claims. Quite how many porn DVD’s may have produced in Germany at that same time I have absolutely no idea – there just aren’t any statistics on the German porn industry to work with – but I think it highly unlikely that it was anything like 400 a week.

That was, however, 7-8 years ago and it’s well documented that economic conditions on the production side of the US porn industry have changed considerable over that period, so to get a more up to date picture I took a quick look at the new releases section of Adult DVD Empire, which is probably the largest online Adult DVD/Video on Demand store in the US, and found that over the last month there have been only 522 new porn DVDs releases into the US market, a little under half the number that were being released each month when the industry was at its peak.

Okay, so you might still look at that figure and think that 500 new DVDs is still a hell of a lot of porn, even after allowing for compilations and scene recycling (i.e. reusing scenes on multiple DVD titles), which was always fairly a common practice in the US industry, but that’s still just a reflection of the fact that the production side of the industry is, and always was, based around small companies producing low budget titles on as fast a turnaround as possible in the knowledge the many new titles would typically sell no more than a few thousands units, including sales to the rental market.

To make any money at all in that kind of market you need a fast turnaround on the production side and that’s why so many porn DVDs are released every month.

There’ll be more on the economics of the porn industry later but, for now, we need to move on to our next statistic:

Internet porn in the UK receives more traffic than social networks, shopping, news and media, email, finance, gaming and travel. (Arthur, 2013)

Okay, so we’re looking here at data supplied to The Guardian by web metrics company SimilarWeb and the data is what it is. How reliable that data might be is, however, a very different question because the article contains an important caveat:

The figures, which do not include traffic from mobile phones…

Ah, that’s a bit of a problem.

In addition to its range of ‘Top Sites’ lists SimilarWeb also has a Top Apps section, albeit that this being beta tested and, right now, covers free Android apps. What that section appears to provide, judging by the inclusion of a number of games in the UK’s Top Apps list, including the ubiquitous ‘Candy Crush Saga’, is a chart based on downloads rather usage or traffic generation.

Nevertheless, at the current list of Top Apps for the UK, three of the top five are social media apps (Facebook Messenger, Whatsapp Messenger – which Facebook has just bought – and Facebook itself) and those are closely followed on the chart by Skype, eBay, BBC IPlayer and Media Player and Snapchat, all of which are in the current top ten. Instagram, Twitter, Viber, Spotify and ITV Player all make the top twenty on the list  and within the top fifty you’ll also find Netflix, BBC News, Amazon, BBC Weather, 4oD, Blackberry Messenger, Gumtree, Kik Messenger, Argos, Kindle, which links directly back to Amazon, and Paypal.

There are a few notable absentees from the current list, particularly Google’s own ‘big three’ apps – Search, Gmail and YouTube – but as these are bundled with the Android operating system I think we can safely assume their absence tells us only that the data SimilarWeb is capturing relates only to new downloads and doesn’t include data on updates.

That being the case, what we can perhaps reasonably infer from this is that absence of mobile traffic in SimilarWeb’s web metrics is almost certain to lead to the company underestimating the amount of Internet traffic in some sectors, especially social media but also perhaps the amount of news, shopping, email and perhaps even banking and finance traffic because a significant portion of those activities have already moved on to the mobile networks from which the company doesn’t currently get any traffic data.

Okay, so against that observation you might say “Well, what about the mobile traffic going to porn sites?”. Well, to begin with to get access to any kind of adult content on UK networks you need to verify your age with a credit or debit card, which can knock quite a number of pay-as-you-go customers out of the frame, then you have to consider that pretty all the major players in online porn are video-heavy tube sites and web-cam sites which won’t half chew a big hole in your monthly bandwidth allowance if you use them and many of those still use Adobe’s Flash Player to deliver most or all of their video content, which Google’s Android operating system no longer officially supports. You certainly can get Flash working on a mobile phone with some Internet browsers, although not Google’s Chrome, but it can be hassle to set it up. Put all that together and it should be obvious that watching porn on a mobile phone over a mobile is not that attractive a proposition compared to what you get on the desktop or via Wi-Fi, so that tends to keep the traffic levels down.


Several recent studies have found that teenagers around the world report using porn to gain information about real life sex (Lauzus et al, 2007) (Wade et al, 2005) (Flood, 2009) ( Giordano & Ross, 2012).

Was it ever otherwise?

If you’re anything like my age then sex education at school consisted of two biology lessons only one of which had anything to say about humans and everything else you had to pick from the reader’s letters sections in shoplifted copies of Fiesta and Knave.

Seriously, young people use a wide range of sources to obtain information about sex and that certainly does include porn. Whether and to what extent this influences young people’s attitudes and behaviours is a complex question which is beyond the scope of this particular articles, which focusses on the accuracy (or otherwise, and it is most otherwise) of Dines’ ‘facts and figures’ but what I will say is that if young people are getting the wrong idea about what real life sex is like from watching porn then that is only because, as a society, we’re doing a piss poor job of educating them about real life and real life sex.

Whatever else porn may or may not be responsible for it ain’t the often shockingly poor state of sex and relationships education in schools and colleges, which no amount of prohibition or porn blocking is going to fix.

Many studies have reported various findings, but across the board it has been found women watch less porn than men.

Well yes, as Ogas and Gaddam note…

On the Web, men prefer images. Women prefer stories. Men prefer graphic sex. Women prefer relationships and romance…

When men and women are free to search for anything they want behind the anonymity of their computer screen, they don’t just seek out different interests. They seek out different modes of stimulation. Men prefer to watch, women prefer to read and discuss. This fundamental dichotomy in sexual interests confirms the predictions of one of the most influential sex scientists, Donald Symons.

That doesn’t mean women don’t use porn at all.

A recent (2013) study of 4,600 Dutch young adults aged 15-25 by Gert Martin Hald and others found that 44.6% of young women included in the study population had used some kind of sexually explicit material over the previous 12 months compared to 88.2% of young men; and of those women just under 70% had watched hardcore porn at least once.

What is perhaps most interesting about Hald’s study – which I must get around to blogging properly – is that after controlling for a wide range of confounding factors he found that across three different behavioural domains; ‘adventurous sex’, ‘partner experience’ and ‘transactional sex’ (i.e. whether or not a young person had ever paid for sex or been paid to have sex with another person) the use of pornography accounted for between o.3% and 4% of the variation in sexual behaviour across these three domains.

That tends to suggest that, at least in The Netherlands, the impact of using pornography on the behaviour of young adults ranges from very small to negligible, but we do need to cautious in what we take from that last observation because we are also, of course, looking at data from a country which is widely regarded as providing its young people with amongst the best and most comprehensive sex and relationship education anywhere in the world.

Again, I don’t want to drift too far off topic so I’ll leave any discussion of Hald’s paper for another time, where I also sort out a copy of the paper for download, but for the purpose of this article the thing to take from it is that yes, men tend to watch a lot more porn than woman but there are limits to what you can actually infer from that statement.

Condoms are only utilized in 10.9% of top rated scenes (Bridges and Wosnitzer, 2007).

Again, the ongoing and complex debate about condom use in porn films is rather beyond the scope of this article, although a little further on Dines cites some other statistics from this same study which I will be delving into in more detail. As statistics go this one is what it is and the citation is correct.

10% of adults admit to having an addiction to online pornography (IFR, 2006)

Ah yes, another Ropelato ‘statistic’ for which no original source was given when it was first published back in 2003 and like many of Ropelato’s other ‘statistics’ it can take quite a while to track down an original source but at least, on this occasion, we have one and its turns out to be an online ‘Cybersex Survey’ conducted by MSNBC back in June 2000.

So our actual statistic here is that in 2000, 10% of a self-selecting unrepresentative group of 38,000 or so Americans who could be bothered to fill in an online survey on a news website claimed to be ‘addicted’ to online pornography based on their own self-diagnosis and straight away we’re in low reliability territory, although but a new paper which sets out to review the entire porn addiction model offers a much better insight in to the problems inherent in the vast majority of claims made about porn addiction.

The prevalence of VSS [Visual Sexual Stimuli] problems reported is inconsistent. Clinicians frequently cite “up to 6 %” of the US population is sexually addicted. This estimate comes from clinical speculation in a popular book [Cairns, 1991] in which the clinician/authors focusing their practice on these issues do not clearly define the boundaries of this diagnosis. Empirical estimates from nationally representative samples are that 0.8 % of men and 0.6 % of women report out of control sexual behaviors that interfere with their daily lives [Skegg et al. “Perceived ‘out of control’ sexual behavior in a cohort of young adults from the Dunedin Multidisciplinary Health and Development Study”. Arch Sex Behav. 2010]. If one assumes these individuals might seek treatment, 82 % of treatment seekers report problems with VSS, and clinicians agree that they have a clinical problem in about 88 % of cases [Reid RC et al. “Report of findings in a DSM-5 field trial for hypersexual disorder”. J Sex Med. 2012]. Thus, VSS problems might affect 0.58 % of men and 0.43 % of women in the USA.

Ley, Prause & Finn. “The Emperor Has No Clothes: A Review of the ‘Pornography Addiction’ Model“. Curr Sex Health Rep. 2014

This new review paper is well worth reading and makes some extremely interesting and illuminating points which are very relevant to the wider debate surrounding Internet pornography, for example:

Many have attempted to generalize the patterns related to problem substance use to explain other behavior problems, including use of VSS. Surprisingly, a clear, falsifiable theoretical model of ‘porn addiction’ has yet to be described. Some use addiction interchangeably with other labels such as hypersexual disorder (HD) – also known as sexual addiction. Others define addiction broadly to refer to any substance or behavior with evidence of excessive appetite: “appetitive behaviour is excessive, at least in the statistical sense”. Simply because a behavior is appetitive and frequently engaged does not mean the behavior is a problem, let alone an addiction. Even when consequences, distress, or dysfunction follow such behaviors, interaction with third variables, such as relationship status or culture, must first be examined.

Setting the sex/porn addiction therapy industry to one side for the moment, many of the biggest promoters of the notion of sex and porn addiction in the US tend to be Churches and Christian organisations for whom any use of pornography, inside or outside marriage, is seen to be inherently problematic, making religious beliefs one of the more obvious third variables that need to be examined when dealing with claims about sex/porn addiction, a point that the same paper picks up on a little further on:

Some have cited personal religious values as providing a conflict between their VSS use and feeling unable to stop. Religious conflict was the main reason cited for problems viewing VSS in one study [Twohig et al. “Viewing internet pornography: for whom is it problematic, how, and why?” Sex Addict Compul. 2009]. Those who want treatment for sex addiction are also more likely to be members of organized religion and hold strong religious values [Winters et al. “Dysregulated sexuality and high sexual desire: distinct constructs?” Arch Sex Behav. 2010 and Ross et al. “Prevalence, severity, and correlates of problematic sexual internet use in Swedish men and women. Arch Sex Behav. 2012]. However, the reverse was not true: religiosity explained little variance (3 %) in the decision to use VSS [Wright PJ. “U.S. males and pornography, 1973-2010: consumption, predictors, correlates.” J Sex Res. 2013]. Far more people report a feeling of inability to control their VSS use, than actually report life difficulties resulting from their use [Skegg et al. 2010]. Feeling unable to stop may reflect personal value conflicts with normal VSS use. No data currently support the notion that ‘porn addicts’ have difficulty inhibiting their VSS use.

Reading the review paper the picture that emerges is one which is entirely characteristic of the practice of disease mongering, in which clinicians progressively widen the diagnostic boundaries of illnesses and, particularly, psychiatric conditions and engage in ‘public awareness’ campaigns in order to expand the market for whatever treatments they have to sell, and make no mistake, the sex/porn addiction business is a lucrative one:

The treatment of pornography and sex addiction is a lucrative, largely unregulated industry. The industry makes many claims for treatment and success, with little (to no) published data. Many treatment centers in the USA have emerged claiming to treat sex addiction. The first 20 inpatient facilities advertising on the internet to treat sex and/or porn addiction in the USA were contacted. They averaged a cost of US$677 (SD = $403) per day. They required or recommended between 9 days to 9 months minimum of inpatient stay. For example, one center claims their sex addiction treatment is “clinically shown to produce results that are up to 3 times faster and 11 times more effective than traditional treatment methods”, although none of the articles on their website (nor in the literature) actually test sex addiction. The use of medications ‘off-label’ to treat ‘pornography addiction’ also appears common. Drugs originally designed to treat alcoholism, depression, and ED have all been suggested. This therapeutic opportunism is well characterized. Some have advocated for transparency, requiring therapists to inform patients that such therapies are experimental, and have not been tested for sex addiction.

Many of the treatment centers and providers also claim religious affiliations, raising questions about the nature of supposed pathology if it is rooted in a particular religion. Some of the most outspoken advocates for an addiction pathology model have publications making explicitly religious arguments against VSS viewing. Religiosity is one of the strongest (negative) predictors of problems with internet VSS use. The risk of conflating profit motive and diagnosis in a population vulnerable due to their strong religious beliefs appears high.

Quite – and ‘therapeutic opportunism’ seem a very apt phrase, so I’m nabbing that one for future use.

And with that it’s time to move on again to Dines’ next claim.

Every 39 minutes a new porn film is created in the United States.

Okay, this yet another figure based on a ‘statistic’ supplied by Jerry Ropelato who, in the December 2007 iteration of his porn stats page, included a table showing the overall growth in the number of US hardcore porn title released each year from 1988 (1,300 new titles) to 2005 (13,588 new titles) using figures obtained from Adult Video News.

So for once we do actually have an identifiable source, even if I had to rely on several other sources to pin the figures to AVN due to Ropelato’s ‘chocolate teapot’ approach to listing his claimed sources, and if you divide the number of minutes in a calendar year (525,600) by 13,588 you get 38.68, so yes, a new title every 39 minutes is about right.

That was, however, eight years ago and the DVD production side of the US industry, as we’ve already seen, has contracted substantially so that currently there appear to be less than half the number of new titles released each month that there were in 2005 making this a claim that is now badly out of date and in need of retirement.

20% of American men admit they access pornography at work.

And we have another one that Dines’ has cribbed from Ropelato and another that traces back to a survey conducted via MSNBC’s website, although in this case we have to go all the way back to 1998.

Cooper, Putnam et al. (1999) found that the prevalence of cybersex activity at work was striking in their recent survey of 9,177 Internet users. Six out of 100 employees reported that their primary method of accessing online sexual material was via their work computers. In total, an amazing 20% of men and 12% of women are using their work computers for at least some portion of their online sexual activity.

Cooper, A., McLoughlin, I. P., & Campbell, K. M. (2000). Sexuality in cyberspace: update for the 21st century. CyberPsychology and Behaviour, 3(4): 521-536.

SAN JOSE, California (CNN) — An online survey indicates that while men are the predominate users of sexual Web sites, young women surf them in higher numbers than expected.

During March and April, some 13,500 people answered 47 questions designed to probe their sexual habits on the Internet. They survey was written by Alvin Cooper, clinical director of the San Jose Marital Services and Sexuality Centre. Cooper analyzed the results of roughly 9,000 questionnaires.

CNN, June 14th 1998.

Need I explain the problem with that one?

And while we’re on the subject of pre-Millennial Internet porn statistics.

70% of all Internet porn traffic occurs during workdays (9am – 5pm) (Sex Tracker, 2012)

Nope, that reference should probably read “(SexTracker, 1999)” although even that could be a little over optimistic.

Pornography, in particular, is a growing concern. Several national surveys and statistics point to the prevalence of e-porn in the workplace:

• According to SexTracker, porn sites receive more than 27 million hits per day, with top sites receiving a staggering 2.8 million hits per day.

• 70 percent of all Internet porn traffic occurs during the 9-to-5 workday, according to SexTracker. This means that one in five workers access cybersex at work.

Internet Usage and the Workplace: It’s Now an HR Issue“, Websense White Paper, November 1999

Do people even use words like ‘cybersex’ anymore?

And the next one…

In the U.S., 2/3 of human resource professionals have found porn on employees’ work computers. (Paintbottle, 2013)

Again, we have a genuine but aging statistic and another citation that took a bit of tracking down – although there is a considerable amount of irony here as the infographic that Dines is citing as her source for her statistic was created last year to promote a new start-up ‘designer’ porn website, which already appears to fallen irrevocably down the Internet memory hole without anyone even noticing.

But as regards our actual 2 out of 3 stat, for that we have to go all the way back to October 2003 and an online poll conducted by a couple of HR websites.

Two out of three human resources professionals have found pornography on their employees’ computers, and for a majority who’ve found such material it has happened more than once, according to an online poll conducted by sister sites and

Asked, “Have you ever found pornography on an employee’s computer?” 43% of the participants said, “Yes, more than once,” while 23% said, “Yes, once.” The remaining 34% answered, “No.”

The two sites drew a total of 474 votes in the poll, conducted the week of Oct. 6.

The results explain why a growing number of employers–worried about liability for their employees’ computer activity–are turning to electronic monitoring. Yet employers also worry about violating their employees’ privacy rights, according to Alexandra Gross, legal editor for Business & Legal Reports, the compliance publisher that operates and

And the next one…

A recent study found that seven out of ten youth have been unwillingly exposed to pornography in the United States (Carroll et al, 2008).

This is one paper I haven’t yet managed to get hold of (yet), so I can’t comment on the specific claim that Dines’ makes here but what is clear is that even if the paper does state that it’s not a finding that the authors considered to be worth including in the paper’s published abstract.

Results reveal that roughly two-thirds (67 percent) of young men and one-half (49 percent) of young women agree that viewing pornography is acceptable, whereas nearly 9 out of 10 (87 percent) young men and nearly one-third (31 percent) of young women reported using pornography. In addition, associations were revealed between pornography acceptance and use and emerging adults’ risky sexual attitudes and behaviors, substance use patterns, and nonmarital cohabitation values. Implications of pornography use during the transition to adulthood were discussed. Pornography is becoming a prevalent part of life in the United States and many countries around the world. Existing studies provide some insight into the pornography patterns of emerging adults and some of the reasons behind their use of pornography, however, there continues to be unanswered questions. This study examined correlates of pornography acceptance and use within a normative population of emerging adults (individuals aged 18 to 26). Participants included 813 university students, both men and women, recruited from 6 college sites across the United States. Participants completed an online questionnaire regarding their acceptance and use of pornography, as well as their sexual values and activity, substance use, and family formation values.

Throw in the fact that we’re looking at a study of just 813 university students and I think it fair to say that whatever this study says the one thing you cannot reasonably do is generalise from its findings to entire adolescent population of the United States, so at least one aspect of Dines’ claim is a complete exaggeration.

Exhausting as following all this might be Dines’ next couple of ‘facts’ are particularly interesting, and we start with…

Children as young as 11 years old are regularly accessing hardcore gonzo pornography (IFR, 2006).

Now here’s that same claim as it has actually appeared on Jerry Ropelato’s site since 2003.

Average age of first Internet exposure to pornography: 11 years old

Straight away you should notice a distinct absence of words like ‘regularly’, ‘hardcore’ and ‘gonzo’ from the claim that Ropelato has been publishing for the last eleven years. Those are additions made by Dines and they appear elsewhere only where people have taken their information from Dines’ most recent ‘facts and figures’ page. Indeed, if we go back to Dine’s first ‘facts and figures’ page, which was created sometime around February/March 2013 and which appeared on her site until it was replaced by the current page sometime around November 2013 then what we find that this same ‘statistic’ was given as follows:

11 years old is the average age of first exposure to Internet pornography (IFR)

That’s a hell of a re-write in just six months on the back of absolutely no supporting evidence.

The ‘average age is eleven’ claims is one I’ve written about on at least three previous occasions but to quickly recap back in November 2005 Seth Lubove of Forbes Magazine successfully traced the origins of that particular claim from a statement made by a US Senator back through a key US political think tank, an article published by the Boston Globe and website which promotes porn-blocking and other parental control tool, all the way back to Ropelato and beyond to Ropelato’s original source:

“Most of the statistics there have come from literally hundreds of sources, all reputable,” Ropelato insists. He says he got the age-11 item from The Drug of the New Millennium, a book about the dangers of porn self-published in 2000 by Mark Kastleman, a self-professed former porn addict in Orem, Utah, who counsels other porn fiends.

As for where Kastleman got that particular ‘statistic’…

“I don’t remember where I got that from,” Kastleman says breezily. “That is a very common statistic.” And there the trail goes cold.

Last year, I did a bit more digging into Kastleman’s background after the ‘average age is eleven’ claim found it’s way on to the pages of the Daily Mail and discovered that in addition to his self-published ‘confessions of a porn addict’ book which, since its second edition in 2007, no longer includes the ‘average age is eleven’ claim, Kastleman is also the co-founder of a Utah-based online sex addiction ‘clinic’ in concert with a psychiatrist, Randall F Hyde, who amongst other things is alleged to have dabbled in a bit of off-the-cuff ‘gay reparation therapy’ back in his days as a psychiatric intern.

In a follow-up article I then went on to look at some actual research on young people’s Internet usage including the data generated by the relatively recent (2010) EUKidsOnline project at the London School of Economics for which it’s clear that the claim that 11 year olds are looking at Internet porn in any significant numbers is complete nonsense. The evidence from that projects shows that 8% of 11-12 year olds had seen any kind of sexual images online in the previous 12 months, the same percentage that reported having seen sexual images on television, film or video/DVD and just 1% ahead of the number who’d seen such images in a magazine or newspaper – and when I say sexual images here I mean anything from a topless ‘page 3’ style image upwards. As for accessing porn, well just 2% said they actually seen sexual images on a adult ‘X-rated’ website compared to 4% who’d run across a pop-up advert.

Okay, so from that we can conclude that the earliest traceable source of the ‘average age is eleven’ claim is hardly reliable and the claim itself is complete and utter rubbish but we’re still no closer to discovering where this idea that large numbers of eleven year olds are watching porn actually originated.

However, while carrying out some of the background research for this article I ran across a reprint of an old ‘Focus on the Family’ article on a Christian website with an original copyright date of 2002 that may provide a tantalising hint as to the actual origins of that claim:

According to the National Coalition for the Protection of Children and Families publications, “Studies show boys ages 12-17 are among the primary consumers of porn — a major source of their sex education.” In a small survey of Christians at the Focus on the Family Institute and a separate youth group, I found that the average age of exposure to Internet pornography was 15, which is a bit higher than the national average. This is due to the fact that a large number of those surveyed were around 20 years old, and the Internet came along during their teen years. The average age for first seeing a pornographic magazine was around 11 1/2 years old. The most interesting part to me was that, among the males, 64.10% had at some point purposely viewed pornography.

We’ll be coming on to the claim that boys aged 12-17 are among the primary consumers of porn very shortly but for the moment let’s focus on the author’s comments about the small survey he carried out and particularly the reference to the average age  for first seeing a pornographic magazine as being around 11 1/2 years old. Although the copyright date on the article postdates the first edition Kastleman’s book it does perhaps suggest that what Kastleman described as a ‘very common statistic’ when interviewed by Seth Lubove might have its origins in much earlier claims about the age at which adolescent Americans encountered their first ever copy of Playboy Magazine.

Now that makes a lot of sense, not least because there was, right from outset, a lot more to Playboy than just its semi-nude and nude pictorials. The magazine itself was launched in 1953 and as early as 1954 the magazine was already making a concerted play for a measure of mainstream respectability by serialising Ray Bradbury’s classic science fiction novel ‘Fahrenheit 451’ across three consecutive issues (March to May). The idea that people bought Playboy for the articles may be seen as a bit of long-running joke, but then consider for a moment this description of the contents of its December 1968 issue:

It’s December 1968 and you grab a mag at the local newsstand. The table of contents includes the following: A quartet of short stories by Alberto Moravia; a symposium on creativity with contributions from Truman Capote, Lawrence Durrell, James T. Farrell, Allen Ginsberg, Le Roi Jones, Arthur Miller, Henry Miller, Norman Podhoretz, Georges Simenon, Isaac Bashevis Singer, William Styron and John Updike; humor pieces from Jean Shepherd and Robert Morley; an article on pacifism in America by Norman Thomas; a piece on how machines will change our lives by Arthur C. Clarke; an essay on “the overheated image” by Marshall McLuhan; contributions from Eric Hoffer and Alan Watts; an article in defense of academic irresponsibility by Leslie Fiedler; a memoir of Hemingway by his son Patrick; Eldridge Cleaver interviewed by Nat Hentoff; a travel piece by the espionage novelist Len Deighton; and the first English translation of a poem by Goethe.

Find me a literary magazine today that could boast of that kind of intellectual fire-power in just one issue.

At its peak in the late 60’s and very early 70’s, Playboy magazine had a circulation of around 7 million copies per issue, so it wouldn’t come as a great surprise to find that some of those copies found their way into the hands of a few sexually curious adolescents and, of course, it only takes one kid in a neighbourhood to successfully manage to sneak one of his dad’s magazines out of the house or accidentally run across a bit of hedge porn for the pictorials to start circulating around near enough every kid in the area.

Whether or not this might have happened with sufficient frequency to make eleven the average age at which American adolescents saw their first Playboy pictorial is another matter entirely but the first centrefold scenario hinted at by that old Focus on the Family article strikes me as being infinitely more plausible than Dines’ ridiculous claim that 11 year olds are regularly watching hardcore gonzo porn.

Frankly, if you were to mention gonzo to most eleven year olds right now they’d assume you were talking about the new Muppet movie.

Now, let’s get back to the business of 12-17 year olds being the largest consumers of porn, which Dines cites as follows:

Following first exposure, the largest consumer group of internet pornography is boys between the ages of 12-17.

Okay, now that is another claim that Dines’ has sourced from Jerry Ropelato and which, on his site, dates all the way back to 2003, but curiously it’s also one that Ropelato dropped when he updated his ‘statistics’ page in December 2007 in favour of:

Largest consumer of Internet pornography: 35 – 49 age group

Although, ironically, that amended figure still appeared in the “Children Internet Pornography Statistics” table.

Now, although the years ascribed to IFR citation given by Dines on her website is 2006, when the original ’12-17′ claim was still in place, the citations given in her book ‘Pornland’ put the dates on which she accessed Ropelato’s as:

– March 2007 for a set of figures used in her chapter on children (chapter 8).

Even though these sites are also becoming increasingly popular with porn users, with nearly 14 million Internet searches for “teen sex” in 2006, an increase of 61 percent in just two years, and 6 million Internet searches for teen porn, an increase of 45 percent over the same period.

– April 2008 for one figure used in the book’s introduction.

As noted, boys see their first porn on average at eleven years of age, and by then most have computer skills sophisticated enough that they can access any of the sites described above.

– And March 2009 for figures used in a chapter on the economic of the porn industry (chapter 3).

The size of the porn industry today is staggering. Though reliable numbers are hard to find, the global industry has been estimated to be worth around $96 billion in 2006, with the U.S. market worth approximately $13 billion. Each year, over 13,000 films are released, and despite their modest budgets, pornography revenues rival those of all the major Hollywood studios combined. There are 420 million Internet porn pages, 4.2 million porn Web sites, and 68 million search engine requests for porn daily.

So, two of the three occasions on which she professes to have visited Ropelato’s site in her book were after the point at which he amended his stats page to state that the largest consumers of Internet pornography are the 35-49 age group and yet, in 2013, she still managed to go back to a version of the page that pre-dated that amendment to obtain figures for her own website, and all, seemingly, without bothering to check even the latest version of Ropelato’s ‘statistics’ page to see if anything had changed.

That’s rather odd in itself but nothing like as odd as the actual origin of the claim that 12-17 year olds are the largest consumer of pornography, which you’ll also frequently see cited in this form:

A primary pornography consumer group is boys between ages 12 – 17.

However, that’s a misquote because what the original source actually says is:

The most frequent exposure to pornography is reported by adolescents between twelve to seventeen, a finding reported by the Canadian as well as the 1970 Commission survey. While sexual knowledge appears to be acquired at younger ages, it remains unclear what role pornography plays in this “sex education” process.

And the source of that statements is Chapter 3 of Part 4 of the Attorney General’s Commission on Pornography: Final Report, AKA “The Meese Report” which was published in July 1986, six years before the publication of the first ever web page in 1992.

However, there is more to this ‘statistic’ in terms of problems than just its age because the assertion that the most frequent exposure to pornography is reported by that age group is based on just two pieces of evidence, a survey undertaken as part of the 1970 report of the President’s Commission on Obscenity and Pornography, which was set up by US President Lyndon B Johnson in the wake of the SCOTUS ruling in Stanley vs Georgia and a 1985 survey of a nationally representative sample of 1,071 Canadian adolescents and men aged from 12 to over 55 years of age which asked about the frequency with which they viewed “Sexually Explicit Films in Movie Theatres and on Video”.

But what, exactly, is a ‘sexually explicit film’ and how was this term described to and understood by those taking part in the survey?

Having been unable to obtain a copy of the actual research study it’s impossible to be sure, which means we need to entertain the possibility that the term “sexually explicit films” could easily include a wide range of films and videos that were not specifically pornographic.

The early 80s was, for example, the golden age of the slasher movie. John Carpenter might have set the mass market ball rolling in 1978 with ‘Halloween’, which is still by some distance the best slasher flash ever made – but the genre really took off in 1980-81 with the release of “Friday the 13th”, “Prom Night”, “My Bloody Valentine” and one of my personal favourites “Happy Birthday to Me” and it didn’t take long at all for female nudity and sex scenes to become an established part of the slasher formula; and even if you weren’t keen on the slasher format but still liked your horror or sci-fi with plenty of skin you’d have well catered for at the time by everything from “The Howling” (1981), the 1982 remake of “Cat People”, 1983’s “The Hunger”, which is another great movie by the way, or Lifeforce (1985) in which the lead actress Mathilda May spends the entire film naked.

On the other hand, if you’d rather have had your naked flesh wrapped up in something a little more serious by way of a screenplay you could have taken your pick from 1980’s “American Gigolo”, 1981’s “Body Heat” and “The Postman Always Rings Twice” or even 1982’s “An Officer and a Gentleman”, which many people seem to forget was actually given a R rating in the US and so could only be seen at the cinema by under 17s if they accompanied by a parent or guardian.

Nor was the early 80s short on American sex comedies which included everything from Screwballs, Risky Business (1983 – starring Tom Cruise) and Batchelor Party (1984, and starring Tom Hanks) to the 1982 film that was the undisputed ‘daddy’ of the American High School sex comedy genre until the emergence of the ‘American Pie’ franchise at the end of the 1990’s – “Porky’s” which, for a lot of adolescent young men at the time was arguably the ‘rites of passage’ movie.

Throw in the home VCR boom and an extensive back catalogue of sexually explicit mainstream films from the 60s and 70s that was also being released on video for the first time and without much more information on the exact parameters of that Canadian survey you have to question just exactly how reliable the assumption that it dealt only with porn might have been 28 years ago let alone its relevance to Internet porn in the 21st century.

So, after tackling a ‘statistic’ which is nearing its 28th birthday (at least), where will Dines take us next?

88.2% of top rated porn scenes contain aggressive acts.

– In 70% of occurrences, a man is perpetrator of the aggression; 94% of the time the act is directed towards a woman.

– Only 9.9% of the top selling scenes analyzed contained behaviors such as kissing, laughing, caressing, or verbal compliments.

– Open-hand slapping occurs in 41.1% of scenes.

– Sex depicted in porn movies generally focuses on men’s sexual pleasure and orgasm, rather than equally that of women’s

(Bridges and Wosnitzer, 2007)

Okay, so I do happen to have a copy of this particular paper and the statistics that Dines reports are what they are albeit that she quotes from the paper in a rather selective manner; for example the most common ‘aggressive act’ recorded in the study was actually ‘spanking’ which was found in 75.3% of the scenes analysed by the researchers.

However, there is an issue with this particular study which we need to tackle, the problem of defining ‘aggression’ in porn films and, indeed, more generally in studies of media violence and aggression.

This particular study, for example, use the same recording methodology that was used in the 1998 National Television Violence Study, which the paper describes as being the ‘Gold Standard’ for this type of media research, and indeed as recording methodologies go the one used in the NTVS study is a pretty rigorous. However, the published results of that study have also been criticised by a number of media researchers for relying on both an overbroad definition of violence and for paying little regard to the context in which violence is depicted on-screen in arriving at conclusions such as this one:

Violence in entertainment media often has unrealistically few consequences such as characters demonstrating little if any pain in response to violent attacks. Wile E Coyote’s reactions to being hit on the head with an anvil or blown up are typical of this point.

To which the obvious response is… IT’S A FUCKING CARTOON.

Exactly how researchers define violence and aggression in any particular study, and whether or not that definition is contextually appropriate to the type of material being analysed are key considerations when evaluating media content analyses, as the study itself explains:

One of the reasons for the higher rates of aggression in the current study is the choice to define aggression as a purposeful act committed by someone that results in harm to either the self or another. This definition, unlike those used in some prior studies (e.g., McKee, 2005), does not require that the target of the aggression attempt to avoid the harm. If such a definition was used, the study would have found that only 12.2% of scenes (n = 37) contained aggression, as these were the only scenes where at least one of the acts of aggression was met with target displeasure and/or attempts to avoid the aggression.

So, in compiling the descriptive statistics that Dines’ quotes on her site the authors of the study treated both the question of whether or not the behaviours they identified as being aggressive were also consensual and the manner in which the ‘target’ responded to that behaviour as being irrelevant to the question of whether what was being depicted on screen would be interpreted by the audience as violence and aggression or as a bit of harmless consensual kink; but when this was taken into account, as has been standard practice in other analyses of the content of pornographic videos, the number of individual scenes in which aggressive behaviours were recorded fell from 273 out of 304 (89.9%) to just 37 out of 304 (12.2%).

That’s a hell of a difference.

Porn scenes have sexist and racist themes through out. Websites often contain menus where users can select genres of women’s ethnicities, body types, and ages. There are also choices such as “amateur,” “interracial”, and the ever popular “teen” category. Men and women who are anything other than white are represented in stereotypical and demeaning ways.

Wow, so one the big revelations here is that porn sites have taxonomic menus to enable people to browse for content that suits their personal tastes.

Again, the question of stereotyping in porn films is rather beyond the scope of this article, so let’s go look at another statistic.

Approximately 20% of all internet pornography involves children (National Center for Missing and Exploited Children, 2013).

And we even have a link on Dines’ site to ‘key facts’ page on the website of the National Center for Missing and Exploited Children on which that particular statistic is conspicuous only by its complete absence.

It simply isn’t there and as far as I can tell it never has been.

Looking around the Internet, this particular ‘statistic’ is most often claimed to have originated in a 2003 report published by NCMEC under the title “Internet Sex Crimes Against Minors: The Response of Law Enforcement” but if you read that report for yourself you’ll find it too does not include the claim that 20% of all Internet pornography involves children.

So where did this claim originate?

In this case I did manage to track down an accurate reference to the original source of this statistics which explains that it actually dates all the way back to a small study conducted in 1994 that looked at just 150 images posted to Usenet groups and found that 20% depicted naked persons who were – or were claimed to be by the user posting the image – under 18 years of age. That said, it was also reported that most of these images were of a kind you’d typically find in naturist magazines and that none of them depicted any kind of sexual activity between an adult and a child/adolescent or between children.

And we’re almost at the end…

Studies show that after viewing pornography men are more likely to:

– report decreased empathy for rape victims

– have increasingly aggressive behavioral tendencies

– report believing that a woman who dresses provocatively deserves to be raped

– report anger at women who flirt but then refuse to have sex

– report decreased sexual interest in their girlfriends or wives

– report increased interest in coercing partners into unwanted sex acts (Bridges, 2006) (Yang, Gahyun, 2012).

A detailed review of media effects research, as it relates to pornography, is again beyond the scope of this article but it’s fair to say that just about the only behaviour that’s ever been causally connected to pornography is masturbation.

And at long last we get to the last set of ‘statistics’…

The porn industry makes more money than Hollywood.  (US Statistics)

– 13,000 adult videos are produced annually, amassing over $13 billion dollars in profit. By comparison, Hollywood released 507 movies and made only 8.8 billion (Bridges and Wosnitzer, 2007).

They also make more than:

The National Football League, The National Basketball Association and Major League Baseball combined and more than NBC, CBS, and ABC combined. In addition, they have larger revenues than the top technology companies (Microsoft, Google, Amazon, eBay, Yahoo, Apple and Netflix) combined (IFR, 2006).

Before we get into this there’s one or two important corrections to be applied to Dines’ figures. The $13 billion dollar figure is not and never has been an estimate of the adult entertainment industry’s annual profits but rather its annual revenues, which is different matter entirely. Likewise that $13 billion figure, which is in any case rather questionable, is supposedly the combined revenue figure for the entire legal US adult entertainment industry, so it includes not only the revenues from the adult video market but also those from cable and satellite TV (including pay-per-view), magazines, strip clubs and lap dancing clubs, sales of sex toys, etc. and, of course, the Internet.

However, the figures given for Hollywood include only the box receipts from movie theatres and do not include any of Hollywood’s revenues from video and DVD sales/rentals, or from cable, satellite and terrestrial television rights (including pay-per-view), or from soundtrack albums or merchandising or any of Hollywood’s other revenue streams. Put all that back into the equation and in 1999, according to figures from Veronis Suhler cited in this 2001 Forbes article, Hollywood’s annual revenues amounted to $31 billion, which is over twice the most generous estimate for the total revenues of the adult entertainment industry. That same article also put the annual revenues generated by broadcast television at $32.3 billion – so much for the claim about NBC, CBS and ABC – with cable television weighing in at £45.5 billion.

As for some of the other comparators that Dines’ throws into the pot, Microsoft reported annual revenues of $39.79 billion for the fiscal year ending on 30 June 2005 while Apple racked up $13.93 billion for the year to September 2005. Bringing those up to date, in 2013 Microsoft racked up $77.85 billion in revenues, which is a little ahead of Amazon who weighed in with just under $75 billion in revenues but well behind Apple, whose revenue for the year exceeded $170 billion worldwide.

Of course, comparing any of those figures to those given for the total US revenues of the adult entertainment industry assumes that this $13 billion figure cited by Dines’ is correct when the reality is that no one really knows what the industry’s combined annual revenues are, not even the industry itself.

Reliable information on the adult entertainment sector is extremely difficult to come by. With a few notable exceptions (Playboy, Penthouse, Hustler, etc.) most of the businesses operating in the sector are private companies who don’t release any kind of sales or revenue figures, while the business intelligence sector, which supplies most of the data you’ll find in newspapers and trade magazines, only very rarely delves into goings on in most of the adult entertainment sector due to the lack of reliable data, publishing occasional one-off analyses in which much of the information given tends to be at best rather speculative.

For example, one of Jerry Ropelato’s claimed sources, Juniper Research, did publish a white paper in 2005 in which they estimated that global revenues from the delivery of adult entertainment content via mobile phones would grow rapidly from around $600 million in 2004 to somewhere around $1.3-$1.4 billion in 2006 and then on to $2.1 billion by 2009, estimates they describe as ‘conservative’ in a way which suggests that other industry analysts at the time were expecting that market to grow a lot more rapidly than Juniper.

Seven years later, in May 2012, the same company were projecting that the mobile market for porn would hit $1 billion a year by 2015 – so what happened to the ‘conservative’ $2.1 billion that the mobile porn market was supposed to be worth by 2009 or the $1.3-$1.4 billion it was supposed to generate in 2006?

Another of Ropelato’s claimed sources, Kagan Research, did publish a market report early in 2005 in which they estimated that porn had accounted for 40% of the US pay-per-view cable and satellite TV market in 2004, giving a dollar value estimate of $800 million for the year.

By the time that figure found its way on to Ropelato’s site at the end of 2007 it had turned in an estimate of $1.34 billion for 2005 rising to $2.19 billion for 2006 with the addition of revenues from the mobile phone and phone sex markets, but where exactly did those figures comes from?

One possibility is AVN which did publish an estimate of $1.745 billion in revenues for 2006 from the adult cable TV and pay-per-view market in January 2007, but this was after Ropelato’s figures made their first appearance on his site and the same AVN press release said nothing at all about the phone sex market and put the value of the mobile market at just $39 million, which is near enough loose change compared to the other figures – and remember Juniper Research reckoned the mobile market would be worth well over $1 billion a year at this point.

The same press release also lays claim to 34% growth rate in the cable TV market over the previous year (2005), which gives a dollar figure for that of a touch over $1.3 billion to set against Ropelato’s 2005 figure of £1.34 billion for cable TV plus phone sex plus mobile traffic, so revenues from phone sex and mobile were just $40 million in 2005 but shot up to over $440 million in 2006? Then again, for porn’s cable TV revenues to hit $1.3 billion in 2005 from Kagan’s 2004 estimate of just $800 million would have required growth of 62.5% in revenues in a single year to get up to AVN’s figures.

Is that plausible> Maybe it is, but then how do we square all that with other figures that were kicking around in early-mid 2007 which showed that Playboy’s US TV revenues were down 16% on the previous year to go with an estimated 28% drop in adult pay-per-view sales on cable networks and figures attributed to Kagan Research which, at best, put the total value of the adult video on demand market on cable TV and direct broadcast satellite networks at around $905 million for 2006 and maybe less than that as that last figure relies on VoD over cable generating $515 million in revenues when figures cited elsewhere, and also attributed to Kagan by Crain’s Chicago Business in March 2007, estimated that VoD over cable TV would only generate annual revenues of $437 million in the US by the end of 2007.

The discrepancy there could, I suppose, indicate that the $905 million and $515 million figures includes international as well as US revenues – the article on from which that figure comes doesn’t indicate exactly which markets those figures relate to – but even so it’s a hell of jump from there to AVN’s $1.745 billion figure for 2006.

Bear in mind here that the cable and satellite TV markets are one of the few areas where it’s possible to obtain financial data on the adult entertainment industry on a regular basis and from independent sources because it, and magazine publishing, are the only market sectors in which there are both a significant number of publicly listed companies and independent auditing of sales figures and buy rates; and yet it’s still not possible to square up the different published estimates found in the trade/business press to come up with a definitive figure for the size of those markets.

So what chance does anyone, even AVN, have of producing reliable estimates for other markets where that kind of data just isn’t available?

Could the adult entertainment market really have been worth as much as $12-$13 billion in 2005/2006?

Looking at the various figures I’ve seen for different market segments I’d have to say that it’s not completely impossible but it is, I think, unlikely as some of the figures given in the various breakdowns stretch credibility further than other. Could the adult magazine market, for example, really have been worth anything from $950 million to $1 billion a year when the market leader, Playboy, were reporting annual revenues for its publishing division of just $25 million a year all-in – i.e. including both sales and advertising revenues? That strikes me as being rather questionable.

Then again, does it really matter if the US adult entertainment industry was worth $4 billion a year, or $8 billion, or $10 billion or even $13 billion a year back in 2006. That’s still eight years ago and the media landscape has changed so much in those eight years that none of those figures will be remotely relevant today.

The same goes for the raft of facile comparators that have been thrown up over the years in conjunction with claims about the size of the adult entertainment sector.

The claim that porn revenues exceeded those of several top tech companies are another Ropelato ‘innovation’ dating back to the 2007 update to his ‘statistics’ page, while the comparison to the NFL, NBA and MLB made its first appearance in the New York Times in 2001 and comparisons to Hollywood’s box office revenues go back at least as far as a 1997 article by Eric Schlosser which was published by US News and World Report. As for the comparison with broadcast television revenues for NBC, CBS and ABC, the earliest citation that I’ve seen for that one claims that it dates all the way back to 1993 and attributes it to the Christian conservative pressure group “Focus on the Family” along with the claim that it had based that claim on information obtained from an FBI investigation into the porn industry, although exactly when that investigation is supposed to have taken place is anyone’s guess. It could have been any time between 1986 and 1993 in the wake of calls for a Federal crackdown on porn following the publication of the Meese Report, which itself claimed that porn was an $8 billion a year industry – Schlosser’s article also gives the same figure in his 1997 article but as an estimate of the size of the industry in 1996 – or it may even date all the way to the late 70s when the FBI definitely did carry out a 2½ year investigation into the involvement of organised crime in the production and distribution of porn – a $4 billion industry at the time according to contemporary reports – leading to 45 individuals being indicted for conspiracy and interstate transportation of obscene materials.

Do any of those figures bear any relation to the financial position of the adult entertainment industry today?


As fascinating as the economics of the adult entertainment industry might be it’s time to wrap up and try and draw a few conclusions; and the first and most obvious thing to say is that it’s pretty clear that, contrary to its title, Dines’ “Facts and Figures” page is desperately short on facts while most of its figures are woefully out of date and/or a product of creative accounting by people with a vested interest in exaggerating the scale of the adult entertainment industry to suit their own ends.

Now you could arguably respond to that observation by pointing out that Dines is by no means the only academic to be completely taken in by Jerry Ropelato’s extremely dodgy ‘Internet porn statistics’ collection nor the first to cherry pick her ‘facts and figures’ from various sources to suit her own personal agenda and promote a particular pet theory, but then no plausible combination of gullibility and cherry picking is going to explain how Dines got from “11 years old is the average age of first exposure to Internet pornography” to “Children as young as 11 years old are regularly accessing hardcore gonzo pornography” or the inaccurate and easily debunked claims she makes about the amount of search traffic looking for ‘teen porn’.

For me that one set of ‘statistics’ rather sums up this entire exercise.

It seems abundantly clear that Dines, or someone working for her, went to a fair bit of time and trouble to come up with what is actually reasonably accurate estimate of the daily number of Google searches for the phrase “teen porn”, more accurate certainly than most of the figures on her ‘facts and figures’ page – it’s not a difficult calculation to make as such but it is an easy one to get badly wrong without a fair amount of care and cross-checking and a reasonable of familiarity with the ins and outs of website metrics.

However, what cannot be reasonably be explained in terms of a genuine error is the leap from that estimate (500,000 searches per day) to the claim that it amounts to around a third of the total daily search traffic for pornographic websites. Seriously, did no bother to carry out even the most basic checks, like maybe running ‘teen porn’ against just the keyword ‘porn’ in Google Trends to see how that one panned out? For the record, the current ratio between the two is 30 to 1 in favour of ‘porn’.

This is a point that I really do think needs to be drive home, so although I’m trying to wrap things up I am go to throw in one last piece of analysis for you to chew over and this of another claim about teen porn that Dines makes in the same self-penned article in Counterpunch in which her other claims made their original appearance:

We also analyzed the content of the three most popular “porntubes,” the portals that serve as gateways to online porn, and found that they contained about 18 million teen-related pages – again, the largest single genre and about one-third of the total content.


Assuming Dines can actually read a top sites list properly then the three most popular porn tubes she should have looking at are XVideos, XHamster and Pornhub.

XVideos currently claims to host just under 4.4 million porn videos and to be adding more than 3000 new videos a day. In reality, the maximum number of videos that can accessed on the site at the moment is a little under 425,000 – it’s browseable index runs to 21,240 pages with 20 videos per page. Exactly how many of those videos are categorised or tagged on the site as teen videos it’s impossible to say – it doesn’t matter whether you use the site’s search facility or its ‘tags’ page, the maximum number of videos it will return at one time for any kind of search or category is just 3,500 so short of manually working your way through 21,240 pages of video thumbnails and counting anything that looks like it might be a ‘teen’ video there is no reliable way of estimating the amount of teen-related content on the site. XVideos does have a page that lists 2001 user generated tags in use on the site with a figure next to each one that ostensibly shows how many videos have had that applied to them, so I can give you a basic headcount for the total number of tags (33.8 million) and the total number of teen-related tags (2.43 million based on a count for every tag that might indicate a teen video) but these figures seem to bear no relation whatsoever to the amount of actual content on the site. For example, the three most used tags are ‘hardcore’ (1.21 million), ‘blowjob’ (1.18 million) and ‘amateur’ (973,000) on a site with only 425,000 videos.

Unlike the other two sites, XVideos doesn’t carry it’s own picture galleries. The pictures tab on the site takes you to a sister site (XNXX) which carries the same video content as XVideos alongside a rather outdated TGP images section in which all the image thumbnails lead off site, mostly to pay sites offering a small number of still images or short video clips in an effort entice visitors to pay for content.

XHamster, the number two site in terms of traffic, is the largest of the three sites with just over 600,000 videos and 931,000 image galleries with an average of around 27 images per gallery, which works out at 26.7 million pages in total, including index pages, gallery page and individual pages for each video and image. Of those, 21.54% of all videos and 20.44% of all image galleries are tagged as teen-related content giving a total of around 5.47 million pages.

Pornhub is the smallest of the three sites with just 112,000 videos and 25000 image galleries again with an average of around 27 images per gallery. 20.2% of the video content is tagged is categorised as teen-related but only 7% of image galleries fall within that category so we have a total page count for the site of 817,500 pages or so of which 72,600 feature teen-related content.

Putting all that together, what we have on these three sites is a little over 28,4 million pages containing either thumbnail images or actual content of which around 5.5 million display teen-related content plus however many teen videos there might actually be on the XVideos site which, based on the other two sites is unlikely to amount to more than 20% or so of the actual video content so maybe around another 105,000 pages. However you look at it, that’s a lot less than the 18 million pages of teen content that Dines claims to have found on those three sites and before you get too caught up in the actual numbers (1.138 million videos and 956,000 image galleries containing over 26 million images) its as well to remember a lot of the video content on porn tubes tends to be short 2-7 minute promotional video clips posted by pay sites and there’s also a lot of duplication of video content via both user uploads and through pay sites reposting the same video clips every couple of months to keep them current. Porn tubes are really no different to any other searchable websites, unless someone browsing the site is looking for something very specific then most won’t bother looking past the first half a dozen pages, so a lot of the content on these sites at any given time is pretty much redundant in terms of the number of visitors who actually bother to look at it.

How do you find 13 million pages of ‘teen porn’ that just aren’t there?

Getting back to wrapping things up, it should be remembered that the starting point for this article is the concerted and, in many respects, desperate attempt that Dines made last year to try the wreck the public reputation of a new academic journal long before it had even published a single paper, one in which Dines and her camp followers even resorted to setting up an online petition attacking the journal for its alleged bias:

In the interest of academic integrity and thorough critical inquiry, it is imperative that a journal titled Porn Studies creates space for critical analyses of porn from diverse and divergent perspectives. Our hope is that you will change the composition of the editorial board, confirm the journal’s commitment to a heterogeneous interrogation of the issues embedded in porn and porn culture, and ensure that diverse perspectives are represented – on the board and also in the essays published in the journal. Failing that, we ask that you change the name to reflect and make evident the bias of its editors (Pro-Porn Studies) and create another journal which will represent the position of anti-porn scholars and activists and the voices of mental health professionals, porn industry survivors, and feminist scholars whose analyses examine the replication and reification of misogyny, child abuse, and sexual exploitation in mainstream pornography (for instance, Critical Porn Studies).

And the sponsor of this petition is..?

Stop Porn Culture and concerned academics

After hacking through her ‘facts and figures’ I’m sure you can imagine what I think of the sight of Gail Dines, who appears to be incapable of properly checking even very basic things like facts and original sources, making an appeal to ‘academic integrity and thorough critical inquiry’ – never mind ‘Critical Porn Studies’, I think ‘Risible Porn Studies’ might make for much more apt title for any future porn journal that would willingly take material from Dines and present it as credible scholarship.

As for the new journal, it remains to be seen what kind of research will actually emerge from it as most of the papers in the first issue focus more on discussing and defining the kind of ground that’s likely to be explored in future issues. If there’s one paper in the current issue that piqued my own interest more than most, not least because of its relevance to some of the issues touched on in this article, it’s a paper by Mazières et al. which some of explores the analytical possibilities offered by the growth of large scale porn tube sites and their use of taxonomic categories and user tags. It’s probably a bit geeky for most people’s tastes but I found it interesting and it hints at other possibilities and research avenue if, for example, porn tube site can be persuaded to open up things like traffic data to researchers.

One of the big questions that emerged for me as I was researching this article and, in particular, looking at the various fragmentary bits of economic and financial data that can be traced with a fair amount of time and effort is that of the extent to which the production of commercial porn films since the low budget video revolution in the mid 1980s may have been driven to a large extent by a relatively small number of regular consumers with quite specific niche interests, creating a rather distorted picture of public tastes in pornography, one that anti-porn activist like Dines go out of their way to exploit in pursuit of their own prohibitionist ends. That’s the kind of question that could be tackled very effectively via detailed traffic analyses using data from the large tube sites to identify not just what content those sites carry, which again, in the case of sites that allow user uploads, may open to distortion by relatively small number of users with niche interests and large personal porn collections, but also what their users search for and, more importantly, actually spend time watching.

Whether any of the major porn tube would be prepared to play ball with that kind of research, as Dogpile did when they supplied their search traffic to data to Ogas and Gaddam, remains to be seen but it’s an avenue that’s worth pursuing and one that can only be pursued by researchers who, unlike Gail Dines, are prepared to approach the subject with an open mind and a genuine commitment to academic integrity and critical inquiry.

6 thoughts on “Gail Dines: Short on Facts, Crap Figures

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.