Here & Now

Noon Wednesday: Tracking Safer at Home

Marisa Wojcik:

Welcome to Noon Wednesday. I’m Marisa Wojcik, a multimedia journalist with Here & Now on PBS Wisconsin. So it’s been about seven weeks since Wisconsin’s safer-at-home order went into effect, to slow the spread of COVID-19 in the community. So how well is Wisconsin adhering to the shelter in place order? Well, researchers have been able to look at location data to track our movement. And it’s showing what patterns have been like after nearly two months of staying at home. So joining us today is Will Cushman from our partners at WisContext. He’s written all about this. Will, thank you so much for joining us.

Will Cushman:

Thanks as always, Marisa.

Marisa Wojcik:

I want to start by asking exactly how are researchers tracking our movement?

Will Cushman:

Yeah, that’s a really good question. And I think a question a lot of people might be wondering, like, “How would they know where I’m going?” And the answer is, your smartphone. So I think a lot of people, probably a majority of people who own and use a smartphone, use it for some, at least one app that tracks their location, at least while they’re using that app. Think about applications like Google Maps or other navigation apps, ride sharing apps, MapMyRun or other running or exercising apps, all sorts of apps will track your location actually, social media apps. And so there are options for allowing or disallowing that location tracking. But I think a lot of people who use those apps for the reason they’re intended are allowing that location tracking to happen. And so what’s happened since a number of I mean, pretty much the whole country went under some, most of the country went under some measures similar to Wisconsin’s safer-at-home order, a number of app developers and then also third party vendors who buy all this location tracking data from app developers to use it for all sorts of reasons, often like targeted advertising or things like that, they have kind of leveraged that data to try to better understand the movements of people kind of before and after orders such as Wisconsin’s safer-at-home order took place. Basically, just to try to understand how much, as you said in the intro, how much people are adhering to these orders, and whether the pandemic is having an effect and how large that effect is on our daily lives and our movements. And what they’re finding is it’s having a pretty big effect.

Marisa Wojcik:

And so it can track our specific location, can it track how far we’ve gone, how frequently we’re moving, and then beyond that, does it track specific demographics? Does it know that it’s me individually going to the grocery store a few blocks away?

Will Cushman:

That’s a really good question. And yeah, I should clarify that it’s anonymized data. So no one knows that Marisa Wojcik is walking to the neighborhood grocery store, or that Will Cushman is going for a run in the neighborhood, or not that I’ve been doing that, but anything like that. It’s all anonymized and it’s all aggregated data. So researchers are able to make claims at a large scale about the movement of people. And I say large scale, but they can actually drill down to pretty local levels. So in Wisconsin, I looked at data that got all the way down to the county level, and I’ve seen data elsewhere that drills down to like zip code-level data. I didn’t get that fine-tuned myself with the data I was looking at. I was looking at data provided by Google itself. It’s the same data that Google uses for maps and like when if you’re looking up on your phone, like, I want to go to Target right now and like Google might tell you if it’s really busy or not, it’s the same, and they can tell you that based on the same type of data. And then I also looked at data provided by a company called Cubic. They’re a third-party vendor, so they have contracts with app developers who track people’s location. And so they have all this aggregated data and they package it out usually for advertising reasons, but they’ve actually put together a pretty handy dashboard on a website that’s publicly available. You just have to create a username and password, yeah, username and password, it’s free though. And that’s what I did. And you can see down to the county level, how far people in specific counties are going, and this is people like at a broad level. So generally people in Dane County, how far they’re going, or how much they are not moving. And so most of the data I was looking at says it’s about, they consider staying at home being staying within 100 meters of one location in any day.

Marisa Wojcik:

And so overall, are people staying at home?

Will Cushman:

Yes, yes. Overall, I would say, you can see a pretty noticeable spike in the proportion of people who are staying home across the state, after the safer at home order went into effect. And even a little bit before that, in mid-March, I think as people really started to understand the severity of the pandemic, and kind of the how serious the situation was. And as schools began to close, you could see, even before the safer at home order, more people staying home, and so that’s more people staying within 100 meters or 330 feet or whatever, are at one location in that day. And across the state, about twice as many people are staying home, or were staying home like right after the safer at home order, as compared to right before the safer at home order. And that varies around the state quite a bit. There’s quite a bit of variation between counties, but generally about twice as many people are staying home.

Marisa Wojcik:

And are people experiencing quarantine fatigue as time goes on? Is that data changing over time?

Will Cushman:

That’s a really good question. And the answer is yes, for sure. I mean, I don’t know about you, but like I’m definitely getting stir-crazy, I think a lot of people are seeing the weather get nicer, the leaves come out on the trees and are like “Oh my gosh, “are we really going to have to like stay literally “in our homes for weeks to come?” And it seems a growing number of people, based on the data, are kind of becoming a little bit more lax in their adherence to the stay at home order. So looking at the data over time from when the order went into effect on March 24, we looked at it like right through the end of April. You can see a pretty steady, small but steady decline in adherence to the stay at home order across the state. Whether it was in a county such as Dane County, or Ozaukee County, which has really high adherence to the stay at home order. Even there, people have been venturing out a little bit more or so the data shows that situation in counties where people were maybe adhering to the stay at home order even less to begin with.

Marisa Wojcik:

How far are we going and where are they going? And what do we know about kind of the difference between different counties if we’re able to see that information that closely?

Will Cushman:

A lot of–basically, like, unless people are staying home, it seems that the data show that the people who aren’t staying home are generally maybe sticking to their previous habits or routines. I think that would make sense just thinking about like, most of the people who probably are not staying home right now are in some way, shape or form, like an essential worker or something so they still have to go to work. And so we’re seeing a number of people, like close to about half the population depending on where you live, still kind of sticking to their typical routines and venturing sometimes pretty far from home on any given day. But we also looked at data, as you mentioned, that can tell us where people are actually going. And that data shows that not surprisingly, people are visiting retail stores and restaurants a lot less. I mean, a lot of restaurants and retail stores are closed right now. I guess like some restaurants are open for takeout or delivery or drive through, but definitely people are visiting those types of establishments a lot less right now, like 60% less than before the stay at home order went into effect. People are also visiting grocery stores and pharmacies less than they were before, not to the degree of the drop off that we’re seeing in visits to restaurants, which makes sense. People still need to get groceries and go to the pharmacy, but it seems like they’re doing it less. One thing I thought was really interesting that the data showed was many more visits to parks after the stay at home order compared to baseline typical conditions. So it seems like a lot of people are interested in getting outside right now which I mean, makes a lot of sense. People are getting a little stir-crazy.

Marisa Wojcik:

How can this information be helpful to public health officials or lawmakers, as a lot of decisions are being made in real time?

Will Cushman:

I think it could be helpful in a number of ways. First, it can just be helpful to, at kind of high, broad level, understand whether or not people are adhering to public health orders such as the safer at home order. So it can just be helpful to gauge how willing the residents of the state are to kind of like play along with these new rules that are really kind of upending lives. And to see kind of this geographic variation in where people are doing so or maybe are able to do so. And I think the location tracking data is also becoming very interesting to researchers and governments who want to leverage it for contact tracing. That is not happening in Wisconsin yet, but it is definitely happening in a number of nations. In East Asia, it’s happening a little bit in Europe, and I know that the state of Utah has released like a beta version of an app that tracks your location. So this, I should be really clear, using location tracking data for contact tracing is totally different than how the location tracking data that I looked at. And that’s because, using it for contact tracing, that data is not aggregated, it’s not anonymized. And it’s coming from people who are actively opting into an app where they’re saying “I know” that this app is going to be tracking my location “wherever I go, including like down to potentially” the level of like, which aisle I’m in in a grocery store. “But I’m okay with that.”

Marisa Wojcik:

Wow.

Will Cushman:

Yeah, you can get like really, really fine-tune with this location tracking, using Bluetooth and in some places, but that’s a whole another story. But I think people are becoming interested in it for contact tracing because as I think a lot of people are experiencing right now, staying home, I think there is a growing realization that we can’t go on like this forever. And at the same time, the risk of the disease remains out in the community. So like what other strategies are there to create as safe an environment as possible. And one of those is to really beef up contact tracing. That requires a huge amount of human resources. So, researchers are pretty interested in how data like passive data collection can maybe make it a little bit easier to contact trace. Obviously, there are a number of privacy tradeoffs and any app like that would have to be opt in here. You just wouldn’t be able, I mean, you just can’t force people to allow a device to track their locations and for a governmental authority to know where they are. But one interesting thing I thought from talking to a researcher at UW who’s doing some of this work, he said that such a location tracking app, if it were to be effective for use for contact tracing, would require 60% of the people in any city or state to opt into it. So that’s a pretty high order.

Marisa Wojcik:

It’s interesting because Google tells us how many people are at Target at any given time or how busy a particular store is and maybe in the future, Google will also be telling us how many people who tested positive for COVID-19 have passed through that location. It’s just fascinating to think about the possibilities of what’s happening so quickly right now. So does any of this information lend to any correlations between positive cases and is safer at home working, or is that not even something that health officials are looking at right now?

Will Cushman:

I don’t think I can say with confidence like whether the data I looked at can show whether like adherence to the stay at home order, whether it’s like higher in some place and that place maybe has a slower rate of infections or a lower number of confirmed COVID cases. Like, I wouldn’t be able to make that claim with any type of like, scientific significance or statistical significance. But just anecdotally, some of the places with really high adherence to the stay at home order, as I think I mentioned a little bit ago that includes Dane County, it includes some parts of suburban Milwaukee, so Waukesha, Washington County, and really Ozaukee County. All those counties have been at or very near half the population staying home, basically, since the safer at home order went into effect. And generally, we’ve seen in those communities slower transmission of the disease. I certainly cannot say whether or not those two things are totally connected. But I think that would be a really interesting hypothesis for an epidemiologist to explore.

Marisa Wojcik:

All right, well thank you so much Will, this is really fascinating. We always enjoy getting your insights on your reporting. So thank you very much.

Will Cushman:

Thank you, Marisa. I always enjoy it as well.

Marisa Wojcik:

And it sounds like if you want to dig into the data yourself, Will said that it’s publicly available. So you can find more of that information and read that entire article at WisContext.org. For more from Here & Now and PBS Wisconsin, you can visit pbswisconsin.org. And thank you so much for joining us on Noon Wednesday.