Prostitution in the US: Services

There is great variation in the services that US sex workers perform

Note: This is the third of several installments. Part 1 explored the dataset, Part 2 looked at the ethnicity of sex workers, part 3 (this part) looks at the services performed and how this correlates to ethnicity, part 4 will look at the economic side of things and how fees correlate to ethnicity, services, looks and performance of sex workers.

Let’s start to correlate

We have a trove of data. We can now start to correlate. To warm things up, I’ll look into each ethnic group’s smoking habits. We need a tiny bit of data cleaning (I’ll turn N/A into “I don’t know”s). Again, this is unlikely to be super-interesting information for most people, but it’s useful to “warm the engine”.

frame.Smokes.fillna("I don't know", inplace=True)
plt.figure(figsize=(10,9))
sns.countplot(y='EthnicityNormalized', data=frame.sort_values(by=['EthnicityNormalized','Smokes']), 
              hue='Smokes');

Smoking habits and attitudes, broken down by ethnicity.

The diagram above tells us something, but the disproportion among the representation of the different ethnic groups makes it hard to compare apples to apples. Let’s see if we can come up with a stacked diagram.

Note: the stacked bar chart wasn’t a native component in Matplotlib nor Seaport, but I really felt I needed it, and I managed to pull it off with the help of this. Integrating it wasn’t immediate though (and I won’t bore you with the details of how I did it), but it did the job in the end.

How smoking habits of different ethnic groups stack up against one another

It’s getting hot in here

So far so good, but now the time has come to look at the more spicy (or distasteful, depending on viewpoints) details. If you are not comfortable with sexual content, here’s your warning that you shouldn’t be reading further.

Is there a correlation between ethnicity and certain sexual acts? Let’s explore fellatio (AKA oral sex, AKA Blowjob, BJ):

plt.figure(figsize=(10,9))
chart = sns.countplot(y='EthnicityNormalized', data=frame.sort_values(by=['EthnicityNormalized','Blow Job']), 
              hue='Blow Job');
chart.set_xticklabels(chart.get_xticklabels(), rotation=90);

Fellatio is pretty much the standard. The happens without condom in about 2/3 of the cases.

Not a ton of differences among ethnicity when it comes to fellatio, with the notable exception that black American providers are more prone to enforcing use of condoms.

Going Further for Your Customers

Time to take a look at how many providers are generally willing to go the extra mile to have happy customers. Many sex workers will let clients come in their mouths. Some will swallow:

frame['Cum In Mouth'].value_counts(normalize=True).plot(kind="barh", grid=True);

24% providers will accept ejaculate in their mouth. Of these, 9% will swallow it

Let’s look at how these sexual mores break down by ethnicity:

plt.figure(figsize=(10,9))
chart = sns.countplot(y='EthnicityNormalized', data=frame.sort_values(by=['EthnicityNormalized','Cum In Mouth']), 
              hue='Cum In Mouth')

Breakdown of the previous chart by ethnicity. We could use side-by-side comparison

This is useful, but comparing differences in behavior based on ethnicity is not immediate. The stacked diagram does the trick. Interestingly, African-American providers are significantly less likely to allow their clients to come in the mouths.

African American providers are generally less likely to accept ejaculate in their mouth (only 4%). European providers are more likely to go for it (around 16%)

For the hell of it, let’s also look at how things work in some major cities:

keycities = ["New York City - Manhattan, NY", "Los Angeles, CA", "Chicago, IL", "Washington, D.C., DC",
             "Mexico City, Mexico", "New Orleans, LA", "Toronto, Canada", "Tijuana, Mexico"]
plt.figure(figsize=(10,9))
sns.countplot(y='City', data=frame[frame.City.isin(keycities)].sort_values(by=['City','Cum In Mouth']), 
              hue='Cum In Mouth');

New Yorkers seem to enjoy swallows slightly more than dwellers of other big cities.

The flip side

Let’s talk anal sex and attitudes towards it by providers with different backgrounds and ethnicity. Latinas enjoy “greek” (a common slang term for anal sex, apparently) more than other ethnicities in the US, but European providers are really open to the idea apparently.

For the overwhelming majority of Asian dolls (K-girls, Chinese, Japanese) out there, in contrast, the rear is no-fly zone pretty much.

Note: YMMV stands for Your Mileage May Very, meaning that some providers may have agreed to providing the greek experience, but their actual performance was lackadaisical.

Latinas appear to be the most open when it comes to anal sex (16%), while Asian dolls are way less willing to indulge (a meager 4%). Western European providers score an amazing 22%

Let’s break it down by city for curiosity.

keycities = ["New York City - Manhattan, NY", "Los Angeles, CA", "Chicago, IL", "Washington, D.C., DC", "Mexico City, Mexico"]
plt.figure(figsize=(10,9))
sns.countplot(y='City', data=frame[frame.City.isin(keycities)].sort_values(by=['City','Anal']), 
              hue='Anal');

Not simple to compare. Let’s stack it:

The Big Easy and San Fran seem to indulge in anal sex a bit more than other major US cities. We added Mexico City and Tijuana for reference. Butt pleasure is more fashionable in Mexico apparently.

Part 3 is over. As you can imagine, the cost of service is a pretty salient aspect in this industry. Correlating ethnicity, geography and the looks with the prices charged for service will be interesting.

That’s what I’ll look at in the next installment.

Part 4: the fees