This week, a Connecticut jury found Richard Dabate guilty of murder in the murder of his wife, Connie Dabate, after a five-week trial that relied – in part – on data from his Fitbit. Richard said a man in camouflage broke into their home in 2015 and shot Connie. But Connie was wearing a Fitbit and data from the device showed movement for about an hour after Richard said the break-in happened.
To prove that the Fitbit data helped show that Richard killed Connie, prosecutors brought in Keith Diaz, an exercise physiologist and professor at Columbia University Medical Center. Diaz has conducted studies validating the accuracy of Fitbits and testified to their accuracy.
Diaz often testifies as an expert witness in criminal trials. He said The edge that it is personally rewarding to participate in trials but that it is also a challenge. The way he typically thinks about Fitbit data as a scientist is different from how he’s asked about it in a courtroom. “The scientific questions we answer are different from the criminal questions,” he says. “What I have tried to do in these cases is to translate that.”
Fitbits are generally accurate devices, Diaz says, but they’re by no means perfect. For scientific research, this is normal – nothing is ever absolute, and it is rare to get an answer with 100% certainty. Science lives with a certain amount of statistical error. But the law operates under different guidelines: it wants to know if something has happened beyond a reasonable doubt. “It has very serious implications if someone could walk away for 25 years for life,” he says. “So, is it acceptable to live with a gray area – and under what conditions? That was part of the challenge and the hurdles to overcome.
Diaz spoke to The edge about his experiences testifying in trials and how he navigates tensions between science and the legal system in the process.
The following has been edited slightly for clarity.
What’s the difference between what a scientist might want to know about a Fitbit and what a jury might need to know?
In science we care about things – like I take 100 steps but the device says I only took 92. It matters in science but how much does it matter in criminal cases ? With these, it’s often just a matter of whether the device detects motion. That’s enough for the deal. It doesn’t matter that it’s 100 steps versus 92; it’s really pretty much if they moved. With this Connecticut case, there’s been a lot of back and forth in cross-examination on the error rate. But the error rate was based on the number of steps someone had taken – not the error rate if they were moving. It’s a very different story.
And in science, especially in many studies that I work on, you work with large sample sizes that could be thousands of people. All noise dissipates with so many people. But, with a criminal case, you only have one person – so the noise is important and can cast doubt on what really happened. In criminal cases, there’s a level of precision that you want to have, okay, that moved at exactly that minute. Compared to science, if there is an error because you have such a large sample, it can be filtered out. So it’s been hard to figure out how you convey that these devices can’t perfectly tell you what’s going on.
How do you adjust the science to meet the needs of a criminal case?
You can take advantage of things that science might consider errors. So, if you are brushing your teeth and wearing a device on your wrist, there are instances where the device has registered it incorrectly as steps. Obviously, in science, this is a mistake. But it can actually be very useful in some of these cases, as it shows that it is a high sensitivity device. It is actually a strength to use these devices in this context because of their high sensitivity. When no movement is detected, when it does not register any steps, we should be very sure that a person is not moving.
So it had to change what we looked at in science to say, “Well, what’s the most important thing in the context of this criminal case?” We can say that this person was no longer moving because the device was recording nothing. So what does that mean in this context?
In a criminal trial, there are two parties, and you are asked to testify for one party or the other. How is it as a scientist?
I was hired by prosecutors, never by the defense. And some prosecutors made efforts to bias my interpretation. It may not be intentional, but they will feel like they believe that person is guilty and they need someone to come out and support the case. But I don’t just want to come in and serve as ammunition for the prosecutor; I want to give a balanced interpretation of what science means and how it can be applied in the situation. It might be easy for me to come in and say things that would support the prosecutor’s case. But how do you balance that with also acknowledging that there’s a mistake and it’s not a perfect device?
It’s hard because obviously the prosecution does not want to highlight the errors. But on the defense side, I noticed that they weren’t asking the right questions about it. They are not scientists, so they don’t know what to ask to highlight the limitations of the device. And I can only answer the questions put to me.
Has this experience changed the way you think about your research?
If it’s something I’m investigating, I would like to conduct more scientific studies. In a lot of situations that I’ve seen with these cases, the science that exists isn’t really designed to answer the important questions. Like: is it possible for a person to walk from a bed to a bathroom 15 feet away without the device registering? So I might look at things like that because I think it would enhance my ability to give expert testimony. And for a jury, it would be useful to have better answers to these kinds of questions that arise in a criminal case and which do not arise in science. But for now, it’s kind of a side concert. It’s a bit outside of my daily work.