Skip to main content
ClearingHouse Community Menu ≡

The Witness in Your Pocket

By Nate Vogel & Michael Hollander mobile phone with pushpin location

Victims of wage theft often face a fundamental barrier to bringing a legal claim to recover their unpaid wages: a lack of reliable evidence. For almost three quarters of a century, the landmark 1946 U.S. Supreme Court case Anderson v. Mt. Clemens Pottery has enabled workers themselves to be the source of reliable evidence in wage cases. As long as a court finds their testimony reasonable, workers are able to testify as to the number of hours they worked or the amount of pay they received. But worker testimony is often not enough: workers’ memories fade, can be inconsistent, or may not be sufficiently reliable where a worker had widely varying hours over a long period of time.

One potential solution to the evidentiary problems that remain post–Mt. Clemens Pottery is cell phone location data.

How? Virtually every adult in the United States has a smartphone.1 Our phones have a variety of sensors on them, such as the Global Positioning System (GPS), that allow them to know where we are at any given moment. For the more than half of Americans with an Android phone, that location information is captured and stored by Google and is available for download by the Google account holder.2 Setting aside the obvious privacy concerns of having Google, application makers, and your service provider know where you are at all times, these data have a huge potential upside for workers: a phone carried to work every day has unwittingly been turned into a neutral schedule maker, recording every moment that the phone is at the workplace. The difficulty is turning that location history into usable evidence.

As detailed below, Community Legal Services of Philadelphia, in conjunction with Hack4Impact, created a web-based tool to turn cell phone location history into evidence.3 The tool transforms raw data into an exhibit that summarizes a person’s time on the job. By carefully applying discovery and evidentiary rules, location data can be persuasive and independent support for workers’ demands for their rightful wages. This tool has the potential to change how wage-theft cases are handled by lawyers, ushering in the next 75 years of protecting workers.

In a World of Information, Still More Needed Sometimes

Aldo’s story is a typical one for many workers in the low-wage economy: he was hired into what at first blush appeared to be a very good job that turned out to be a wage-theft trap.4 Aldo was hired to work for a contractor undertaking a massive renovation of an old hospital that was being converted into high-end condos, a gym, and a restaurant. Not only was the pay reasonable at $120 a day, but also the project was well financed by a New York developer and was scheduled to last for at least two years. Two years of regular pay without fear of not being paid seemed like a great job to Aldo.

The job did not turn out how Aldo imagined, and he ultimately left the job after a year and a half. Seeking help in recovering his last few weeks of pay, Aldo came to Community Legal Services. We quickly realized not only that Aldo was owed his last few weeks of wages but also that he was likely owed an unknown amount of money in overtime pay. Aldo reported regularly working nine hours a day, five or six days a week. He sometimes worked for days on end, sleeping on the job site.

Despite having worked more than 40 hours in most weeks, he was always paid $120 per day. If he worked five days, he received $600; if he worked six days, he received $720, regardless of the number of hours that he worked. Under both state and federal wage laws this is illegal.5 Instead he should have been paid 1.5 times his normal hourly rate for his hours over 40. If he had worked 45 hours over five days, for example, he should have been paid $633.35, not $600.6

Wage Theft

Wage theft is nothing new and has been written about extensively.7 Broadly speaking, wage theft is any one of a number of ways that employers fail to pay their employees properly.8 Wage theft likely costs workers over $50 billion per year.9 Common forms of wage theft include complete or partial failure to pay a promised wage, off-the-clock work, pay at less than minimum wage, and a failure to pay overtime for hours worked over 40. Local, state, and federal laws protect workers from wage theft and give them recourse if they are victims of wage theft.

Despite the strength of these laws, however, a problem remains: lack of information.

Lack of Information

The lack-of-information problem is simple: if workers do not know how many hours they worked in a given week, they cannot figure out whether they are being paid properly. Relatedly, if workers cannot testify in court as to the approximate number of hours worked in a given week, they cannot prove liability or damages.

This problem is not easily solvable. A worker who had the same schedule for weeks or months at a time may have a good sense of hours worked (“I worked about 45 hours a week.”), but what about a worker whose hours vary daily or weekly and who will not be able to give anything more concrete than vague estimates (“Sometimes I worked 45 hours a week; sometimes I worked 60 hours a week; and sometimes it rained so we only worked 20 or 30 hours.”)? A worker could keep a record of hours worked, but rarely do workers think to do this ahead of time. Employers may keep discoverable records, but many do not, especially in the low-wage economy.

Mt. Clemens Pottery

In Anderson v. Mt. Clemens Pottery Company the U.S. Supreme Court tackled this very issue: in the absence of a reliable employer-maintained record of a worker’s hours, how can a worker prove liability and damages?10 The Court found that in the absence of accurate employer-kept records, an employee needs to proffer only “sufficient evidence to show the amount and extent of [his damages] as a matter of just and reasonable inference.”11 The burden then shifts to the employer to show more precisely the hours worked or to show why the employee’s evidence is not reasonable.12

In other words, an employee can present whatever evidence will allow the court to infer reasonably the employee’s hours. In the 72 years since Mt. Clemens Pottery, workers have presented evidence in the form of the worker’s testimony, a worker-created daily calendar of hours, statistical studies, and virtually every other form of evidence imaginable.13

Not Always Enough

Consider Aldo: He worked for a year and a half; his hours varied on a weekly basis; and, like most workers, he did not keep a running log of his hours. Looking back a year and a half later, he had no way to know how many hours he worked in any given week and thus no way to know what he was owed by his employer, let alone prove it to a judge.

Some workers in this situation can reasonably approximate their hours by testifying that they generally worked a certain schedule that varied on a semiregular basis (e.g., 45 hours a week, with an extra nine-hour day of work every fourth week). Aldo, however, had a widely varying schedule: he worked as few as 25 hours in some weeks and as many as 90 in others.

What Aldo did not realize is that he had the perfect witness: his cell phone.

Big-Brother Location Tracking in a Pocket-Sized Format

Now that cell phones are ubiquitous, consumers can easily forget how miraculous a cell phone is. This pocket-sized brick of glass, plastic, and metal allows us to reach out to anyone we choose, at any time, anywhere in the world. If the cell phone is also a smartphone, we can also send emails, update social media, and pay bills. Applications for the phone can enable us to sign a mortgage or pay taxes while sitting on a beach.

Cell phones are great for consumers, but for companies and governments they offer incredible additional features. The most important of these for the purposes of this discussion: a cell phone is effectively a tracking device that you have decided to wear. Location data can be immensely valuable: for example, companies can target ads based on proximity to products, and governments can locate suspects. Since the widespread adoption of cell phones, millions of people have accepted (consciously or not) keeping a tracking device with them at all times and to pay for the privilege of doing so.

Legal scholars, engineers, journalists, and many others have written a great deal about how cell phone location tracking works, and so we will keep our discussion of the technology superficial.14 The key concept is that in order to operate, a cell phone must send and receive a variety of different signals with a variety of third parties, including cellular signals from antennas on buildings, GPS signals from satellites, and even Wi-Fi signals from local Internet access points.15 None of the technologies for locating mobile devices is completely accurate. The signals they rely on bounce off buildings, trees, and traffic.

But an entity—a company, a government, or anyone else—that can obtain records of these signals can estimate, with varying degrees of accuracy, where the cell phone that sent the signals was located.

Millions of people have accepted (consciously or not) keeping a tracking device with them at all times and to pay for the privilege of doing so.

Suppose you are a friendly technology company with lots of customers with cell phones. You want to know the locations of these cell phones so you can—well, let us just say that you want to offer better services to your customers. You have access to the cell tower, GPS, and Wi-Fi signal information on your customers’ cell phones, but none of these is totally accurate. What are you to do? Which location technology do you use?

The solution that Google has chosen, and the solution that probably many other companies have chosen, is “all of the above.”16 Thus a party such as Google with access to your phone’s records combines cell tower, GPS, and Wi-Fi signal information to come up with a more accurate estimate of your location than it could using only one of these technologies.

The location estimate is still not perfect. Many factors, such as a phone’s battery efficiency or movement, can lead to errors.17 But imperfect location data can still be very impressive (or frightening, depending on your perspective).

Perhaps recognizing that many people might feel uncomfortable knowing that Google stores extensive records of their location, Google allows users to see and use the location data that it has collected on them. Seeing Google’s collection of location data about you is straightforward. Simply visit www.google.com/maps/timeline, and log into your account. Unless you have managed to avoid Google’s location tracking net, you will see a map with dots indicating your locations at different times.18 For example, we visit Harrisburg, Pennsylvania, fairly often to meet with other Pennsylvania civil legal aid advocates. Google knows all about it (see fig. 1).

Harrisburg map
Figure 1. A map taken from Google's timeline tool shows where one of the authors has spent time in Harrisburg, Pennsylvania. Map data: ©2018 Google.

You can also download your data in a text-based format.19 The downloaded location record contains a list of Google’s estimates about your location. Each item in the list represents Google’s estimate of your location at a particular moment in time. The data have a simple structure. Below is an example:

------------------------------------------------------------------------------------
{
  "timestampMs" : "1526064356533",
  "latitudeE7" : 399504870,
  "longitudeE7" : -751653558,
  "accuracy" : 15,
  "activity" : [ {
  "timestampMs" : "1526064360064",
  "activity" : [ {
  "type" : "STILL",
  "confidence" : 99
}, {
  "type" : "UNKNOWN",
  "confidence" : 1
 } ]
}
 }
-----------------------------------------------------------------------------------------

The “latitudeE7” and “longitudeE7” values indicate where Google thinks you were.20 A timestamp identifies when Google believes you were there. The data also include an “activity” attribute that estimates what you might have been doing when the estimate was made.

Note that one of the attributes here is “accuracy.” Google predicts only that you are within a circle around the estimated location and that the radius of that circle is of a certain number of meters.21 In my dataset, Google’s accuracy ranges from a few meters to hundreds of meters.

The widely ranging accuracy of Google’s location data is a useful reminder. These devices we carry around with us are incredibly powerful. They enable location tracking of consumers on a scale that would have been unimaginable a few decades ago. But our cell phones are not magic, and we are capable of understanding their potential and using them to serve our clients in new ways.

Using Cell Phone Tracking Information to Win a Wage-Theft Case

The creepy part about cell phone tracking—that our phone always knows where we are—is exactly what makes it so valuable in a wage-theft case. For those workers who regularly carry around an Android smartphone and have location tracking turned on, their phones have created location histories that can stretch back for years without any prior effort by or knowledge of the workers. With the right tools, that location history can be turned into a work schedule, which, under Mt. Clemens Pottery, can be used in court to prove liability and damages. Because that location history is linked to a Google account rather than to a specific phone, the location history persists even when the worker switches to a new phone.

A number of features of cell phone location tracking make it ideal for a wage case:

  • Location tracking is often turned on without the worker realizing or intending it. This means that data collection began long before the worker realized the worker had a legal problem that required evidence.22
  • Location tracking is neutral and stored independently. Because the information is generated by a phone, spoofing is unlikely; because it is stored by Google, the underlying data are not subject to manipulation.
  • Location history is reasonably accurate. With some exceptions, it should be seen as a trusted source of information of a worker’s whereabouts.

Once made into a work schedule, phone location history can be primarily used in two ways under Mt. Clemens Pottery:

  1. To corroborate a worker’s detailed story. When a worker has a precise recollection about the worker’s schedule, cell phone location history can be a neutral, contemporaneously recorded information to bolster the worker’s story.

  2. To supplement a worker’s general story, supplying details where the worker’s memory is vaguer. When a worker lacks precise recall as to hours worked over the course of months or years, cell phone location history can be used to supplement more general testimony with detailed information. The worker may testify that the worker regularly worked between 40 and 60 hours a week, without being able to say the frequency of any given number of hours. The worker’s cell phone location data can then give precise numbers of hours for each week, ideally coinciding with the worker’s testimony.

The Basic Math

As detailed above, our phones regularly record our precise latitude and longitude, often many times a minute. Those data can be easily (from a computing standpoint) manipulated to determine when a phone was at any given location over a period of time.

Consider a simplified set of six points of data from a location file:

Time Latitude Longitude
7:00 a.m. 30 30
7:05 a.m. 32 32
12:00 p.m. 32 32
12:05 p.m. 32 32
5:00 p.m. 32 32
5:05 p.m. 35 35

Using simple geometry, we can construct a line from point to point. That line would move from 30,30 to 32,32 between 7:00 a.m. and 7:05 a.m. It would stay in the same place until 5:00 p.m., when it would move from 32,32 to 35,35.

If we imagine that our workplace is a building marked by the coordinates 31,31; 31,33; 33,33; and 33,31, we can see that our line starts outside our building, moves into the building, and then moves out of the building again (see fig. 2).

data point plot
Figure 2. A simplified plot of data points from a location file can show when a worker entered and left the workplace.

To construct a schedule, we simply have to say that we entered work each time the line crosses into our building and that we left work each time the line leaves. In the example above, our schedule would show that we entered work at 7:05 a.m. and left work at 5:00 p.m.

A Google location file can easily have thousands of entries in a day, but the math remains the same, just on a different scale. In order to simplify vastly the processing of a Google location file, Community Legal Services worked with Hack4Impact, a nonprofit, undergraduate group at the University of Pennsylvania, to create a website that allows a user to upload location data, mark the location of a worksite(s), and ultimately construct a work schedule in a CSV (comma-separated values, i.e., Excel-readable) format.23

The website uses the same basic algorithm as above: it looks at each successive location in the Google location history and compares it to the last point. If one point is “at work” and the other is not, it marks the time that the phone entered or left work and then processes the next point. By processing all of the points in this manner, the program is able to mark down every time the phone entered or left the worksite—in essence, a work schedule.

Using the Location-Tracking Website

The Wage-Theft/Cell Phone website is simple to use:

  1. Download your location data from Google.

  2. Load your location data into the website.24

  3. The website plots all of your locations on a map. Points are clustered together and spread out as you zoom in. The initial map may show a single circle covering the eastern United States; when you zoom in, that circle spreads out to a number of smaller circles throughout the region (see fig. 3).

  4. Draw a box around your worksite and give it a name. If you have multiple worksites for the same employer, you can put a box around each worksite and give each a name.

  5. The website then calculates all of the times when you entered or exited one of the boxes and creates a file that can be opened in Excel.25

Philadelphia map showing data points
Figure 3. A map with location data clustered together. Map data: ©2018 OpenStreetMap contributors, CC-BY-SA. Imagery: ©2018 Mapbox.com.

The output is a spreadsheet with seven columns: location, start date, start time, end date, end time, duration, and total time. Each row of the spreadsheet represents a time period spent at work—generally a full day or part of a day if the worker left for lunch. A single day at work may be represented by one line in the spreadsheet or many lines, depending on how often the worker left the worksite during the day. Each week then has a total number of hours for the week.

Below is a small sample of output from the spreadsheet.

Name of Area Start Date Start Time End Date End Time Duration (hrs) Total Time for Week (hrs)
Work 6/6/2018 8:51 6/6/2018 17:21 8.432289167  
Work 6/5/2018 14:09 6/5/2018 17:19 2.6082475  
Work 6/5/2018 8:57 6/5/2018 12:56 3.924338611  
Work 6/4/2018 9:10 6/4/2018 17:14 7.701628333  
            22.66650361
Work 6/1/2018 15:09 6/1/2018 17:53 2.706888056  
Work 6/1/2018 12:21 6/1/2018 12:50 0.474808333  
Work 6/1/2018 9:03 6/1/2018 12:04 2.948841111  
Work 5/31/2018 12:23 5/31/2018 17:29 5.093651944  

 

The advocate will likely want to add columns to help calculate back wages. Before doing this, make a copy of the first tab in the spreadsheet, and do any manipulation on a second tab so that the original data are not touched.

Potential Problems

Recognize that location tracking is not perfect. Every advocate proposing to use this information should know about a few common problems and account for them in constructing a case.

  1. Location tracking itself is not perfect. Particularly in a city, large buildings can obscure GPS signals or cause them to bounce. And at times location history is based solely on the nearest cell tower. Both of these problems can cause the phone to appear to be up to several hundred feet away from its actual location.

  2. A phone can stop tracking location. This can happen if the phone is turned off (because of a dead battery or otherwise) or if the phone enters an area, such as a deep basement, with no Wi-Fi, cell, or GPS coverage. This will make the phone stop tracking location, and this will shorten the time that the phone appears to be in the workplace.

  3. A phone may not always be with the worker. Location history can track only where a phone is, not where a worker is. If a worker left the phone at home, for example, the cell phone location history on that day would not reflect any work hours. Similarly, if the worker left the phone at work, it would show additional hours that the worker did not work.

  4. Phones do not know when you are working. Cell phone location history can track only where the phone is, not the owner’s activity. A worker may take lunch breaks on-site each day, and this would be recorded by the phone as time “at work.” If a worker takes lunch near work, the worker may also appear to be “at work”; the same is true if the worker lives across the street from work or spends free time at or near work.

All of these problems can be overcome; the potential proponent of cell phone location history need only be aware of the problems and their work-arounds.

Problem 1 can be taken care of by drawing a somewhat overgenerous box around the worksite to compensate for inaccurate signals (which could lead to more of problem 4). Problems 2 and 3 can be compensated for by looking carefully at the work schedule that is generated by the website and spotting any anomalies (such as days with very few or no hours when more hours are expected) and speaking with the worker about those time periods. Perhaps the worker sometimes forgot the phone at home or forgot to charge the phone’s battery. Perhaps the worker was off-site for a few days, and this led the phone to record no on-site work. Problem 4 can be compensated for by ensuring that lunch breaks are removed from any calculations of hours and by declining to use location history for workers who live at or very near their work.

Using Cell Phone Location History to Help Aldo

Cell phone location history can help Aldo recover maximum wages. Without location data, all Aldo can do is estimate the hours that he worked each week. This is allowed under Mt. Clemens Pottery, but if he can give only a general range of 45 to 54 hours worked a week, without knowing how often he did either, he may be stuck winning overtime for 45 hours a week—the only number that he can say with certainty that he worked each week.

If he carried around an Android phone, however, he can use the schedule created by the website from his phone data to bolster his testimony and confirm how many hours he actually worked each week. Instead of settling for 45 hours a week, Aldo can collect his unpaid overtime for exactly the number of hours that he worked each week. In Aldo’s case, this could add nine overtime hours a week to the five overtime hours he already counted. If he worked 54 hours every other week, that adds up to an extra 234 hours per year beyond his minimum estimated overtime of 45 hours per week.

Legal Basis for Using Location History in Court

Conquering the technological complexity of employing cell phone location data to investigate wage-theft claims is only half the battle. We must also be able to use in litigation the information we glean from clients’ phones.

Using this type of information in wage-theft cases is still a new strategy, and so we cannot cover the full range of legal complexities that could arise. However, based on experience and conversations with other experts in the field, we have identified several major legal issues that an advocate must navigate. One set of issues relates to discovery: when do discovery rules require disclosure of location information, and what location-related information must an advocate disclose? Another relates to introducing location information as evidence: how can an advocate introduce location records, and whom should an advocate bring into the case to support introducing the evidence?

Answering these questions also requires distinguishing the different types of records we seek to use in litigation. The raw location data, downloaded from Google, are in a large file of text in which a human cannot find very much useful information. The data become much more useful with a computer program that can parse the data, make calculations, and render a new document that summarizes the raw data. The rules of discovery and evidence treat these different records—raw data from Google and a record of location information that is based on the raw data—differently.

Discovery Disclosures

An advocate who reviews location data and parses them with a tool such as the “What’s my schedule” site will be required to share, during the initial disclosure phase, the raw data and whatever parsed summaries the advocate may use at trial. Federal discovery rules require a litigant to disclose copies of documents in the litigant’s possession that the litigant “may use to support its claims or defenses.”26 A party is not obligated to volunteer information that it has decided not to use, whether harmful or not.27

If an advocate reviews the location data and concludes that the data may be used at trial to support wage-theft claims, the advocate will need to turn over the raw data to opposing counsel.28 (The advocate may disclose a description of the data rather than a complete copy.) Because the client’s data are likely to include a substantial amount of personal information irrelevant to the wage claims, an advocate should consider seeking a protective order and redacting irrelevant portions of the location data before sharing them.29

Raw location data will not be much use at trial without parsed summaries explaining the data in a more accessible format. The first few summaries you create will likely simply be rough notes as you explore the data and learn about how accurate and reliable they appear to be. This process is important for revealing inconsistencies between the data and a client’s recollection. For example, you may discover that on some days a client’s family member borrowed the phone, and the client had forgotten until you discovered the apparent trips to other places when the client is sure the client was at work. Or the client’s location data may fail to give a good enough picture of the client’s work schedule to use at trial. For example, if the client works in a place with no cell service, the location data may have no information about the client’s whereabouts during the critical times. These notes are confidential work product, like any other notes or drafts an attorney makes while working on a case, and this work product will not need to be shared in discovery. However, once an advocate has created a document to be used at trial, in a deposition, or otherwise, the document will need to be disclosed.

Introducing Location Information as Evidence

After making the appropriate disclosures in discovery, an advocate’s next hurdle will be to use location information at trial. The clearest approach will be to introduce location data not as the authoritative source of truth about the client’s location but as corroborating support for a stronger source of evidence: the worker’s own testimony. In the absence of clear worker testimony, the worker can still testify as to the generalities of the worker’s schedule (“I often worked 50-to-60-hour weeks, but I cannot remember the precise number of hours each week.”), and location data can flesh out the worker’s testimony.

Clear, uncomplicated testimony from the worker about the worker’s schedule is always the best evidentiary source. As helpful as location data may be, they remain subject to reliability attacks. Location data are also complicated—they potentially require an expert witness to explain how location tracking works, the contents of the location tracking file downloaded from Google, and the inner workings of the website that translates these location data into usable evidence. Providing clear testimony from a worker is easy to do and lays a strong foundation for the judge to understand later location testimony.

During the worker’s testimony, the attorney will need to lay a foundation for later introducing location data. The attorney should make inquiries along the lines of “Did you have a cell phone during the relevant time periods?” and “Did you keep your cell phone with you when you went to work?” The attorney may want to anticipate and acknowledge limitations with the location data: “Were there any times you may have given your phone to anyone else, not kept it with you, or let the battery drain completely?” Once the client has testified as to the work schedule and missing wages, an attorney can introduce location data to supplement and bolster that testimony.

Decades of federal wage-theft law support a finding that location data have the accuracy necessary to support the plaintiff’s burden to prove hours worked. The Court in Mt. Clemens Pottery held that, in the absence of accurate employer records, a factfinder should “draw whatever reasonable inferences can be drawn from the employees' evidence.”30 The Mt. Clemens Pottery Court allowed a factfinder to rely on evidence that was less than perfectly accurate to infer workers’ hours. Here, too, Mt. Clemens Pottery should permit courts to consider location data, despite the acknowledged limitations.

Once the client has testified as to the work schedule and missing wages, an attorney can introduce location data to supplement and bolster that testimony.

The raw location data should be admissible as a business record—a record that is regularly and contemporaneously “kept in the course of a regularly conducted activity of a business.”31 Business records are out-of-court statements, but they are not barred by the hearsay rules. Location data from a personal Google account are undoubtedly regularly collected records that are an important part of Google’s business operations, and so categorizing raw location data as a business record is unlikely to be a challenge.

A complication arises because federal rules additionally require business records to be introduced by a “custodian or another qualified witness, or by a certification.”32 Subpoenaing a Google engineer to testify about the location data is a complicated and expensive proposition. Instead an attorney may be able to save time and expense by obtaining a certification about location records under Federal Rule of Evidence 902(11) and (13) or even by stipulating the records’ admissibility with opposing counsel.

An attorney can employ the parsed summary as demonstrative evidence to supplement the raw data. The summary may or may not be formally admitted into evidence, depending on how an attorney uses it. The Federal Rules of Evidence distinguish between demonstrative exhibits that merely supplement testimony but are not themselves evidence and summary exhibits that are actually evidence. Rule 611 gives a judge broad authority to allow witnesses to present their testimony, but a demonstrative exhibit used under Rule 611 is merely an aid for the jury’s understanding; it is not technically substantive evidence. For the jury to be permitted to take a piece of demonstrative evidence into deliberations, the evidence must meet the stricter requirements of Rule 1006.33 An attorney must make a difficult choice between these two options.

The attorney will need to introduce the parsed summary through a witness’s testimony, proceeding under either Rule 611 or Rule 1006. Attorneys will need to exercise care in structuring this witness’s testimony. Federal courts distinguish fact from expert testimony, and presenting expert testimony triggers a variety of disclosure rules and other requirements that an attorney may prefer to avoid. The line between fact and expert testimony is not easy to find, particularly in this context.34 The question depends on many factors, including whether the testimony involves personal observations or hypotheticals and whether the court finds the testimony involves knowledge “well beyond that of the average layperson.”35 In practice, especially in local courts, the line can become blurrier still.

Decades of federal wage-theft law support a finding that location data have the accuracy necessary to support the plaintiff’s burden to prove hours worked.

One approach would be to have the witness sit down with the client, download a fresh copy of the location data from Google, and run the data through the summarizing application. This way, the witness can testify to the facts surrounding the summary’s creation, from acquiring the raw data to preparing the courtroom-ready piece of evidence. The witness will not need to offer any opinions about the worker’s actual location or explain the application’s code in great detail (which a court might find is beyond a layperson’s understanding). The factfinder will be left to find the location information persuasive or not.

The witness should have some understanding of how the application works. If an attorney is using “What’s My Schedule” to create a parsed summary of location data, we suggest tasking the witness with examining some of the following resources to learn about the technical operation of the application: (1) this article, (2) Hack4Impact’s article describing its application, published on Medium.com,36 (3) the explanatory text on the live version of the application,37 and (4) the code of the application, which is freely available on the Internet.38

The place of evidence and testimony involving records produced with some form of algorithm is an evolving area of law. Our brief discussion here is inevitably incomplete, and the important boundaries, such as that between fact and expert testimony about the application of algorithms to data, are guaranteed to change with time and jurisdiction. We hope at least that the issues we raise here help attorneys using location data begin the journey through this complex landscape and successfully introduce their clients’ location data at trial.

***

Workers’ rights attorneys seeking to bring wage-theft claims have always had to be creative: employers who are willing to steal wages from their employees rarely keep around the stark evidence that would prove their malfeasance, and workers are rarely prescient enough to keep their own meticulous records (although that does happen). Mt. Clemens Pottery opened the door to introducing all manner of evidence, from worker testimony to daily schedules to handwritten daily diaries on the backs of envelopes.

Location data present an extraordinarily powerful new form of evidence—one that is neutral (it is created and kept by a third party), passively recorded (the worker does not need to start a program each day), and ubiquitous (many workers already carry Android phones with location tracking turned on). This evidence has the ability to create far stronger cases than currently exist for low-wage workers and may transform cases that were too weak to bring into strong cases and cases for lower dollar amounts (because of rounding hours down to the lowest reasonable estimate) into higher-value cases. And, like any good litigation tool, this evidence has the potential to force earlier and better settlements for workers and potentially avoid lengthy litigation altogether.

Like any tool, this evidence also has its weaknesses and drawbacks. Location data are not perfect and can be challenged in court. A worker’s location data may be spotty and not useful in a trial; much like DNA evidence in a jury trial, if factfinders come to expect location data, their very absence may appear damning. We are entering a brave new world, one with great hope and many pitfalls.

Download this article
↑ Go up to the top.