"You're wrong about" podcast and data about human trafficing

"The answer is always more spreadsheets." -Michael Hobbes

The good news: 1) This isn't a COVID example. 2) This is one of those examples that gets your students to think more abstractly about some of the tougher, fundamental questions in statistics. Precisely: How do we count things in the very, very messy real world? What are the ramifications of miscounting messy things? 3) The example comes in the form of the very engaging podcast You're Wrong About, hosted by Michael Hobbes and Sarah Marshall.

@yourewrongabout

The bad news: The example is about human trafficking, so not nearly as fluffy as my hotdog or seagull posts.

That said, this episode of the You're Wrong About podcast, or even just the first 10 minutes of the episode, reveals how hard it can be to count and operationalize a variable that seems pretty clear cut: The number of children who are trafficked in America every year. 

The You're Wrong About podcast takes misunderstood, widely reported events and breaks them down. They've covered controversial topics like The Satanic Panic (a good one for anyone teaching cognitive/false confessions/false memories). The snipers who terrorized DC during the Summer of 2009, the OJ Simpson Trial, Chandra Levy's disappearance. Also, they did an EXCELLENT summary of Jessica Simpson's autobiography. I mean, it is riveting and touches mental health, substance abuse, childhood sexual abuse, bullying, sexism in the music industry. It is a good podcast.

Anyway, the 11/25/19 episode focused on misunderstandings about human trafficking, with an emphasis on the trafficking of children. The podcast starts by explaining how the statistics related to counting the instances of child trafficking are likely incorrect. The podcast then delves into better-documented, verified cases of human trafficking: People being lured to the US with the promise of legitimate jobs, to find out that they are being forced into prostitution. 

Now, there is no doubt that human trafficking is terrible, and that it exists. However, the data that is often cited about human trafficking is poorly calculated, and an excellent example of bad stats.

Here is a link to the episode in question, or you can listen to it via a web browser here.

How To Use In Class:

1. Counting and measuring things may be a lot harder than it seems at first blush.

The statsy portion begins at the 3:19 mark.

1. Where do the estimates of child sex trafficking come from? The hosts talk about a lot of foggy, difficult to pin down estimates of the number of trafficked children. Specifically, they question The National Center for Missing and Exploited Children claim that 1:7 runaways are likely victims of trafficking. 

In 2015 The Washington Post fact-checked those statistics. What did that investigation uncover? If you call the NCMEC hotline, and you imply, or it is understood (by the person answering the phone) that it might be trafficking, then it is counted as trafficking. 83% come from foster care agencies reporting, but parents and friends may also call, and the child can be counted multiple times. Also, if the same foster care agency or parent reports a child missing numerous times (because the child runs away multiple times), that one child can count as numerous instances of trafficking. Also, there is no follow up to see if the child has been found. There is no verification of whether or not trafficking has taken place. 

The host looked into different sources of data to estimate the discrepancy between what is reported and what exists. DHS found 500 victims of trafficking, adults, and children. Meanwhile, NGOs and claiming that there are 10s, if not 100s, of 1000s of victims every year. 

Please note: I think some children have terrible lives of abuse. I do not doubt that. It just seems like this specific form of exploitation is being miscounted and overestimated.

2. Know where your data comes from!

Sometimes, flawed data grows legs and is shared widely without anyone knowing where the data came from in the first place. I wrote a previous blog post about the fact that our belief that the human body temperature is 98.6 is flawed and based on data that is over 100 years old. The human trafficking data is another example of this phenomenon and demonstrates why fact-checking is so very important. 

3. What are the real-life ramifications of bad data?

As discussed by the hosts, imagine if all the time and energy and donations that went to human trafficking was also aimed at well-documented, verifiable sources of harm to children: Poor nutrition, child abuse in the home, educational inequality, crime, etc.? The world would be a better place and more children would have better outcomes. 

4. Bad data doesn't mean bad will. 

There are many instances of large, influential organizations hiding away to influence public opinions and avoid lawsuits. I genuinely don't believe that is the case here. I think that the organizations that are trying to help human trafficking victims are sincere, but their techniques for data collecting need improvement.

5. How could we better approach data collection to track child trafficking in the United States?

This could be a good discussion prompt for your students. They could work through it individually or in groups, in a Zoom break out room or in a GroupMe chat. 

6. What do statsy-data jobs look like in real life?

They have a really good convo about the potential for data to solve problems BUT data can be janky. That starts at the 7:50 mark. A favorite quote of mine: "It is so frustrating to realize that the solution to our big social problems is we need wonks and people who are fixated on the details and want to do the grinding, meticulous work of getting it right." So, the right person for the job has statistics training but also a passion to fight against human trafficking. Similarly, there are many different organizations and industries that need wonks. As such, a statistician can choose a career path that satisfies their desire to work with data but also specialize in a field of interest.