Wednesday, December 7, 2011

Porn, Rape, Hamza Tzortzis and Statistical Analysis Part 2

Last month I wrote a review of Hamza Tzortis' article entitled 'Do Liberal Societies Facilitate Rape via the Legalisation of Pornography?' As a member of a liberal society I took a dim view of this. It's worth reading before going further.

I brought my article to Hamza's attention for his comment. In fairness to Hamza, he read this post, describing it as 'quite thorough', and we discussed for a few FaceBook posts. As always he was polite. I've met him in person and cannot fault his manners or eloquence. He wondered if the figures on Google Insights could have been caused by increased liberalisation or materialism in Muslim countries. He said that rising property prices meant people were marrying later in life, forced to live with their parents for longer and that this could be the cause. This was an off-the-cuff hypothesis on his part, I don't mean to represent it as an entrenched position.

How can we check for this?

My first thought was to cross-reference property price trends in the countries in question with census data on marriage age. Unfortunately census data on marriage age tends to be patchy and gathered maybe once a decade. Also, attempting to conduct a worldwide survey of property price trends for the last decade did not mesh well with my plans for sleeping and washing in my spare time.

Instead I decided to break out the figures from my first article year-on-year.

For those of you who skipped my recommendation to review the first article - I used Google Insights(tm) to run stats on which countries consumed the most pornography. In order to control for differences in legislation from country to country I focused on pornography which is illegal worldwide.

By breaking out these figures year-on-year I hope to give Hamza or any other interested parties the opportunity to find a correlation between a major economic or political upheaval, a sharp rise in property prices or a similar cause and the high rates of pornography consumption in certain countries.

Side note - I've automated the process of gathering these statistics and can make them (or indeed different search terms of your choice) available in tabular format on request. I check comments daily - just let me know what you'd like and a method of sending you an Excel spreadsheet.

Searches for child sex, 2004 - 2011, by country

Searches for horse sex, 2004 - 2011, by country

Searches for rape sex, 2004 - 2011, by country

5 comments:

pippin101 said...
This comment has been removed by the author.
Anonymous said...

A little reading Geoff, would do you some good:

http://archives.dawn.com/archives/120067

Do I sense intellectual dishonesty? Hmm.

Geoff said...

Hi Anonymous,

Thanks for taking the time to comment. For those who haven't read the link, dawn.com is a large, established and well respected English language news outlet in Pakistan. They contacted Google for comment on an unrelated story that Pakistan was leading the world in searches for pornography run by Fox news - not a news source I personally favour.

Google replied by saying: "We do our best to provide accurate data and to provide insights into broad search patterns, but the results for a given query, such as those reported in this story from Pakistan, may contain inaccuracies because the sample size is too small for the results to be statistically sound."

Dawn.com's article makes a good point - Fox said that Pakistan leads the world in searches per person, failing to take into account differences in internet access. Dawn says 5% of the Pakistani population had internet access, but I feel this is inaccurate. The International Telecommunication Union puts it at over 10%, and the Pakistani Telecommunication Authority maintains current and historical stats on internet access. Now to my mind we should also include a chunk of the mobile phone users (68%), but identifying how many have data plans, how many have smart phones and how many use the functionality would be a challenge. Let's just take it as a given that fewer people in Pakistan have internet access than, say, Ireland.

Is this something I ignore? To quote myself:

You can read more about how the data is normalised here. It's not a breakdown relative to population size, though considering the difference in broadband penetration between the two nations this wouldn't be ideal for our purpose. (It would hardly be fair to call Pakistan a lower consumer of pornography just because fewer people have internet access.)

Rather it ranks the searches described against the popularity of other searches. In other words, if you take a list of everything searched for on Google, the terms described are more popular in Pakistan than any other country.

It's not perfect, but I think by checking the top ten countries and checking three unrelated search terms it gives us a good approximation.


To add to that, I feel that breaking out the results year on year shows that the increase in broadband penetration has had little effect on the liberal / non-liberal split when it comes to pornography. And at the risk of stating the obvious, this isn't an article about Pakistan, nor is it the only country mentioned.

MiloBoz said...

This is a dishonest cherry picking article aimed at making muslims look bad.. No Muslim countries come up at all for the following terms:

- Rape Fetish
- Beastiality
- rape porn free
- beastiality porn

That is just random ones I typed in .. There is many more without any, or 1 or 2 muslim countrise out of 10.. You're really reaching with this article...

Geoff Shorts said...

Let's examine the "dishonest and cherry picking" claim first.

You say no Muslim countries come up at all for the following terms:

- Rape Fetish
- Beastiality
- rape porn free
- beastiality porn

for "rape porn free" the top ten features Pakistan, Bangladesh and Malaysia. Why did you say otherwise? Bestiality and fetish are words known only to those with a fairly advanced knowledge of English which is why you can be fairly safe in assuming non-English speaking countries won't feature too highly.

It is you who are dishonest and cherry picking.

This two part series was not an attack on Islam. I made a point of highlighting that liberal Muslim countries like Turkey do not feature on the list. This is a defence of liberalism after Hamza Tzortzis said it caused rape.

If you'd like to make an honest attempt to rebut it, feel free to try again.