ISOM Faculty Research

The NYPD gave officers iPhones. Here’s what we learned about race and policing

Jennifer Anzaldi — Tue, 04 Jun 2024 16:50:43 +0000

The NYPD gave officers iPhones. Here’s what we learned about race and policing Jennifer Anzaldi Tue, 06/04/2024 - 12:50

In This Story

Brad Greenwood

The controversy about biased policing seems to draw endless fuel from race-based differences in public perception. Simply put, the vast majority of White citizens in the United States believe the police are doing a good job, including on issues of racial equality, while a similar percentage of Black citizens hold the opposite opinion. And while a growing number of studies have indicated persistent patterns of racial discrimination in policing, an emergent concern among scholars is that the data these papers rely on are also subject to baked-in biases, since they often derive from officers’ self-reports of their own behavior.

Brad Greenwood

Enter Brad Greenwood, professor of information systems and operations management at the Donald G. Costello College of Business at ��AV. One of his research interests lies in how digital technologies are bringing unprecedented transparency to police practices. For example, Greenwood’s 2022 paper documented how the introduction of body-worn cameras for the New York Police Department (NYPD) resulted in a significant reduction in abuse-of-authority complaints.

His latest work on policing is forthcoming in Proceedings of the National Academy of Sciences. Along with Gordon Burtch from Boston University and Jeremy Watson from the University of Minnesota, Greenwood examined the recent rollout of iPhones across the NYPD, which included a series of digital tools designed to replace the handwritten memo books officers previously relied on. Instead of scribbling in the physical books, which NYPD officers were required to hang onto even into retirement, officers could log their activities directly into a centralized database maintained by the NYPD. These detailed digital records shed fresh light on how cops spend their time—and attention—on the beat.

The researchers tracked data on NYPD stops and complaints in 2017 and 2018, the period when iPhones were being rolled out across precincts in New York City. A curious pattern emerged. There was an 18% increase in reported stops after a precinct received iPhones, which would be consistent with the digital tools making it easier for officers to report a citizen interaction. Further, the researchers discovered that this increase resulted in neither more arrests nor more complaints from the public. It wasn’t, therefore, that the phones were somehow causing the police to stop people more often, but rather that so-called “unproductive stops”—those leading to no further action—were being reported more often.

However, when breaking the results down across White and non-White citizens, the researchers found that unproductive stops involving non-White citizens were entirely responsible for the increase. In other words, the observed changes were based on police encounters with non-White members of the public, that would likely have gone unreported in the days of pen and paper. More specifically, after switching to the smartphone system, officers logged 22% more stops involving non-White citizens, while the number of reported stops of White citizens remained unchanged. These are statistical averages—the pattern was more marked in high-crime neighborhoods and those with a greater proportion of non-White residents.

Greenwood offered an interpretation of the finding: “The concern here is that we have an underreporting, which is concentrated in certain groups and means that we need to be cautious when interpreting prior work. On the one hand, it opens the door to bias in police interactions with civilians being worse than initially anticipated, at least based on the frequency of stops. On the other hand, it could mean that older data doesn’t accurately reflect the likelihood of an arrest once a stop occurs. And we need to be doubly cautious, because we don’t know if officers are reporting stops more frequently just because it is easier, or for some other reason.”

Greenwood cautions against making sweeping conclusions based on the study. “The only thing we know for sure is that more and deeper work is needed by scholars and policy makers to ensure transparency between law enforcement and the people they are charged to protect,” he said.

On the whole, however, the study raises the possibility that race-based disparities in policing are not only very real, but may have been underestimated thus far because of reporting gaps.

As police officers are not obligated to document all civilian interactions, their decisions regarding what—and what not—to report can be biased. The introduction of new technology, as in the case of the NYPD, can help counter such biases, but is not the only avenue worth pursuing. The researchers recommend that police departments “investigate the appropriate organizational complements (i.e., policies and procedures) necessary to uncover and eliminate such biases.”

Topics

impact fall 2024

George Mason faculty are tackling cybersecurity’s talent pipeline problem

Jennifer Anzaldi — Fri, 10 May 2024 17:01:06 +0000

George Mason faculty are tackling cybersecurity’s talent pipeline problem Jennifer Anzaldi Fri, 05/10/2024 - 13:01

If you’re a cybercriminal, the latest news on cybersecurity talent shortfalls should put a smile on your face. For example, the majority of cybersecurity leaders report that their teams are understaffed, and they have problems retaining qualified professionals.

But for Nirup Menon, a ��AV professor of information systems and operations management (ISOM), and Brian Ngac, an instructor in the ISOM area, this workforce challenge is a golden career opportunity for the young people of Northern Virginia and the Washington, D.C., area.

Nirup Menon and Brian Ngac

The pair recently won a two-year award from the National Institute of Standards and Technology (NIST), an agency within the U.S. Department of Commerce, to create unique experiential learning opportunities and workshops designed to enhance cybersecurity education and workforce development.

Working closely with industry partners Mobius Consulting and Institute for Defense Analyses (IDA), Menon and Ngac will recruit and help select students to work on actual cybersecurity projects. “They need to have taken some fundamental cyber class ahead of time,” Menon clarifies. “We want students with a commitment to the field. It allows you to get experience but it’s also competitive.”

Throughout the 12-week projects, students will receive mentoring both from the industry participant and from business faculty. “We run it in an agile scrum-like manner,” Ngac says. “Every week, we ask ‘What did you do?’ ‘What are you going to do?’ ‘What are the challenges that are impacting your work?’” If students run into trouble, faculty mentors can work with industry managers to help them get back on track.

“We’re trying to build not just the cyber workforce but the skills as well,” Ngac says.

Menon and Ngac have developed a specialty in this type of hands-on learning, which they have dubbed the Professional Readiness Experiential Program (PREP). More than 100 Virginia-based undergraduates and 20 industry participants have participated in PREP, which includes projects funded by two Commonwealth Cybersecurity Initiative Experiential Learning grants in collaboration with Mobius and IDA.

“PREP not only focuses on cybersecurity projects, but also works on many business process improvement projects,” says Ngac. "Honors and high-performing ISOM students work on real-world projects with industry participants on identifying technical solutions to business challenges through rigorous research, modelling, analysis, quantification, risk management, implementation planning, and, at times, execution.”

The NIST award also incorporates workshops for students who are new to cybersecurity but interested in exploring it as a career option. Workshops will be launched in collaboration with Trinity Washington University (TWU), an HBCU whose College of Arts and Sciences is women-only. For a field such as cybersecurity, which continues to face diversity challenges, the participation of organizations such as TWU is essential.

“We want to bring in students who have not thought of cybersecurity as a field, because they think it’s all engineering, hacking and coding,” Menon says. The workshops will emphasize the variety of functions that are integral to the space, such as management and auditing, in addition to engineering.

Students and industry participants in the current CCI Experiential Learning Projects

“It’s not just tech, there may be creativity involved in anticipating scams and threats,” Ngac explains. “These are different things we’ll be bringing up in the workshop in terms of roleplaying what cybercriminals might do, or how someone might try to socially engineer an attack.”

Unlike a standard grant, the NIST award is structured as a cooperative agreement in which the funding agency will collaborate in shaping and delivering programs as they evolve.

“The advantage of working with NIST is that top people work there. They are the standards body, so they have seen and surveyed a lot of industry,” Menon says. He also lauds NIST’s high-level view of cybersecurity and its implications. “They’re not just looking at technology but also public policy, human factors, etc. It’s a holistic approach.”

Organizations interested in being an industry participant (whether they have cybersecurity-focused or business process improvement-focused projects) with PREP are encouraged to contact Brian Ngac.

Join the Mason Nation

Topics

Commonwealth Cyber Initiative (CCI)

workforce

Students

National Institute of Standards and Technology NIST

Tech Talent Investment Program (TTIP)

Strategic Direction

Cryptocurrency’s surprising transparency advantage

Marianne Klinker — Tue, 02 Jan 2024 18:18:45 +0000

Cryptocurrency’s surprising transparency advantage Marianne Klinker Tue, 01/02/2024 - 13:18

Despite the fears of regulators and skittish investors, clear and accurate signals of cryptocurrency quality may be hidden in plain sight.

As perhaps befits a product of the post-2008 economy, the cryptocurrency space has never known normalcy. In a mere 13 years, crypto went from an untried software innovation of mysterious origin to being touted as the future of investing by major movie stars in Super Bowl commercials. Soon thereafter, of course, came the “crypto winter” of 2022, which began well before, but was surely deepened by, the downfall of FTX and disgraced wunderkind Sam Bankman-Fried.

Forthcoming in Journal of Management Information Systems, a recent paper is the first to examine whether the intensity of developer engagement with a cryptocurrency could indeed be related to its quality.

The crypto backlash appears rooted in the belief that the sector is too rife with bad actors, and too technologically complicated, to be worthy of public trust. The consensus in the crypto community, however, is that accurate clues to crypto quality are hidden in plain sight. The majority of cryptocurrencies use open source platforms such as GitHub for the development of their software. Seasoned investors will routinely inspect the publicly available development history of a cryptocurrency as a basic aspect of due diligence.

Forthcoming in Journal of Management Information Systems, a recent paper co-authored by Mariia Petryk, an assistant professor of information systems at the Donald G. Costello College of Business at ��AV, is the first to examine whether the intensity of developer engagement with a cryptocurrency could indeed be related to its quality. (Her collaborators were Liangfei Qiu and Praveen Pathak of University of Florida.)

Mariia Petryk

“This paper is about how we evaluate reputation in a setting where traditional reputational instruments are not precise,” Petryk explains. “When we choose a doctor, we often rely on an endorsement from someone we trust. But in lieu of that, we look into the history of the doctor—education, training, etc.” By the same token (no pun intended), crypto investors could “judge quality based on the total effort developers made to write the code.”

Using GitHub data for 559 cryptocurrencies over the period August 2016 to December 2019, the researchers zeroed in on five open-source activities. They classed three under the heading of “quality enhancement.” Commits are code modifications submitted by developers and eventually added to the crypto source code. Before official adoption, proposed modifications exist as pull requests awaiting assessment. Issues are questions, bugs, or problems raised by developers or users for discussion within the open source platform.

The remaining two activities fall under “diffusion of software,” essentially a proxy for developer attention. Forks are mirrored copies of the original code attached to a developer’s account. Watches are a way for developers to “subscribe” in order to more closely follow the discussion within the community around a cryptocurrency.

Across the data-set, the researchers found that a one-standard-deviation increase in forks and watches would equate to a 0.56% price increase per month, or 6.7% per year.

As for the three “quality enhancement” metrics, increases in issues exerted upward pressure on token price, but pull requests and commits displayed the opposite effect. A one-standard-deviation increase in issues led to a 4.3 percent higher price over one year, while an equivalent increase in pull requests was linked to annual losses of five percent.

These contradictory results perhaps point to uncertainties about the length of time it would take for crypto admins to evaluate and implement proposed changes. It goes without saying that greater numbers of requests for revision would heighten such uncertainties and their possible implications for cryptocurrency quality.

Overall, these findings imply a “virtuous circle” whereby more promising cryptocurrencies attract more developer attention, which in turn produces quality improvements that are reflected in the token price. While the expansion of the developer community creates the potential for rising pull requests and commits to lower the token price, the price-positive impact of forks and watches was approximately six times stronger. As a general rule, then, intensity of developer attention and engagement could be viewed as a leading indicator of how the market values the tokens—with the very important caveat that broad statistical patterns may have little to no bearing on specific investment decisions.

Therefore, information from GitHub and other open-source platforms could be an important reference point for policymakers looking to evolve nuanced regulatory approaches. “One purpose of regulation is to create equal opportunity,” Petryk says. “Transparent mechanisms allow investors to learn about underlying assets and properties and make their own judgments. Open source platforms like GitHub are open to everyone, but not everyone thinks to use it. In terms of info disclosure, this could be an important factor.”

More Costello College of Business Faculty Research

In This Story

Mariia Petryk

Are U.S. ‘news deserts’ hothouses of corruption?
November 26, 2024
When CEOs are haunted by memories of past recessions
November 19, 2024
MS in Finance students answer burning investment questions
October 22, 2024
The work-from-home blues have a secret source: nostalgia
September 19, 2024
When expressing gratitude, it’s all in the timing
September 4, 2024

Topics

Do you criticize or celebrate your colleagues? It may depend on your social position

Marianne Klinker — Wed, 27 Sep 2023 14:49:21 +0000

Do you criticize or celebrate your colleagues? It may depend on your social position Marianne Klinker Wed, 09/27/2023 - 10:49

Online technology has made real-time performance feedback a workplace reality. But a pair of Mason professors have found out about a major bias in the system.

Online technology is fundamentally reshaping employee evaluations. In the last decade or so, companies such as IBM, Amazon, and General Electric have adopted performance feedback apps that allow employees to "review" one another in real time. These apps take the 360-degree paradigm to its logical extreme by removing temporal, hierarchical, and geographical barriers to feedback.

According to Mariia Petryk, assistant professor of information systems and operations management (ISOM) at ��AV School of Business, “People are trying to tap into new sources of employee engagement across all management and employment tiers. For millennials and Gen Z, instant communication is the norm, and they are not going to wait a year to get feedback. They want to know how they perform here and now, and be able to comment on other people’s performance in the same way. So when we merge these trends of social connectedness, instant communication, and use of technology, we come up with this wonderful application.”

As with any breakthrough technology, though, appropriate use of real-time performance feedback depends upon understanding its inherent limitations. After all, increasing the scale and speed of feedback is not guaranteed to erase deep-seated biases based upon gender, race, hierarchical position, etc. In a recent paper for Information Systems Research, Petryk and her ISOM colleague Siddharth Bhattacharya, concentrate on a relatively neglected—but, as it turns out, subtly powerful—category of bias related to how individuals are embedded within the informal (i.e. social) network of the organization.

Mariia Petryk (left) and Siddharth Bhattacharya

Their co-authors were Michael Rivera and Subodha Kumar of Temple University, and Liangfei Qiu of University of Florida.

Working from a unique data-set from technology provider DevelapMe, comprising nearly 4,000 instances of real-time performance feedback spanning five organizations, the researchers mapped the informal networks of each organization. They then compared reviews submitted by employees who were positionally embedded—i.e. those who moved in influential circles, though they may not themselves have been high-ranking—to ones by those who were structurally embedded, meaning they had larger clusters of weak ties.

For millennials and Gen Z, instant communication is the norm, and they are not going to wait a year to get feedback. They want to know how they perform here and now, and be able to comment on other people’s performance in the same way.

For example, both a C-level executive and their assistant could be considered positionally embedded. A middle manager whose work touches multiple teams would be structurally embedded.

The professors found that positionally embedded employees tended to give higher scores to colleagues, while structurally embedded employees skewed negative in their ratings.

Bhattacharya says, “Informal network bias could be explained as a matter of perspective. From atop the hierarchy, it’s difficult to see how projects came together and who made what happen. Positionally embedded people have a coarse rather than a granular view. Therefore, they may give highly visible individuals more credit than they deserve for collaborative work—for example, they may wrongly assume that a team member chosen to present a project to them was primarily responsible for said project.

By contrast, structurally embedded employees have wider and more diverse networks and thus a much broader base of comparison. This makes them prone to detect and emphasize the flaws of co-workers.”

Organizations using feedback apps such as DevelapMe can usually set limits on the number of anonymous reviews they allow, although the identities of anonymous raters are always visible to HR and senior leadership. The researchers found that anonymity magnified informal network bias for both structurally embedded and positionally embedded employees.

Further, since reviews included both numerical score and explanatory text, the researchers analyzed how informal network bias influenced the wording of reviews. They saw that positionally embedded raters, though more generous with their numerical rating, were relatively neutral and formal in their written feedback. Structurally embedded reviews exhibited the opposite pattern: Comparatively strict in their scoring, but positive and encouraging in their written content. The researchers speculate this points to contrasting motives—constructive vs. motivational—the two groups had for delivering feedback.

Easy ways to counter informal network bias, then, would be for organizations to carefully consider the amount of anonymity to permit, and for them to recommend or even require that each instance of feedback be accompanied by text.

Beyond that, Bhattacharya and Petryk suggest that companies employ a combination of training and technological refinements to help address informal network bias. For example, positionally embedded managers should be reminded to temper their reviews with a bit more objectivity—perhaps peering outside their bubble to get a more complete picture of an employee’s work. Tech providers and consultants could use tools such as social network mapping to help organizations better account for informal network bias in their employee performance data.

Petryk says, “Our data and findings show the mechanisms of how people—not necessarily high-ranking people—can have power over rewards, because at the end of the day the ratings will be factored into a formal evaluation. And bonuses will be distributed on the basis of the evaluation. How that decision is being made can be greatly impacted by the data that we analyze, and that we obtain from this network.”

Topics

More School of Business Faculty Research

Are U.S. ‘news deserts’ hothouses of corruption?
November 26, 2024
When CEOs are haunted by memories of past recessions
November 19, 2024
MS in Finance students answer burning investment questions
October 22, 2024
The work-from-home blues have a secret source: nostalgia
September 19, 2024
When expressing gratitude, it’s all in the timing
September 4, 2024

In This Story

Mariia Petryk

Siddharth Bhattacharya

When managing creatives, what you say is often what you get

Marianne Klinker — Fri, 30 Jun 2023 11:15:50 +0000

When managing creatives, what you say is often what you get Marianne Klinker Fri, 06/30/2023 - 07:15

A new ��AV study explores the complex connections between managerial feedback and creative outcomes.

The growing popularity of crowdsourcing and other forms of open innovation reflects the pressing need that companies have for creative ideas that go beyond the organizational same-old, same-old.

But once you have imaginative outsiders ready to lend you their time and attention, how do you elicit novel and useful contributions from them? It turns out to be as much about strategic communication as it is about the quality of your talent pool.

In recently published research, Pallab Sanyal, professor and area chair of information systems and operations management (ISOM) at Mason's School of Business, and Shun Ye, associate professor and assistant area chair of ISOM, focused on two types of feedback crowdsourcing participants commonly receive. Outcome feedback rates the perceived quality of the submission, with no underlying explanation (“This design is not good.”). Process feedback reveals or hints at what contest organizers are looking for (“I prefer a green background”).

Pallab Sanyal (left) and Shun Ye (right)

Sanyal and Ye analyzed data from a crowdsourcing platform covering close to 12,000 graphic-design contests over the period from 2009 to 2014. The data-set included the contest parameters, time-stamped submissions and feedback, winning designs, etc. It also allowed the researchers to track the activity of repeat entrants from contest to contest across the sample.

This put them in a good position to measure how choosing one feedback type over the other affected contest outcomes—but not in terms of “quality” as it is traditionally defined by researchers.

“The moral of the story is, beauty is in the eye of the beholder. Whoever is the contest holder or client, whatever they think is best for their business objective, that is the highest quality.” —Pallab Sanyal

“I gave a talk at a university where I showed 25 different submissions from a crowdsourcing contest and asked people to choose which one was the highest quality," says Sanyal. "And everyone in that room picked a different one. Not only that, the one that eventually won the contest was not picked by anyone.”

“The moral of the story is, beauty is in the eye of the beholder. Whoever is the contest holder or client, whatever they think is best for their business objective, that is the highest quality.”

With this working definition in mind, Sanyal and Ye developed an artificial intelligence (AI) tool for scoring all submissions by visual similarity to the eventual winning submission.

“We use the algorithm to calculate the distance between these images and the highest-quality image, to give it a score, a quality score, between zero and one,” Sanyal explains.

They found that process feedback tended to increase the affinity of the designs, i.e., they were more similar to the winning design chosen by the client on average. By contrast, outcome feedback increased the diversity of the designs.

Sanyal and Ye theorize that precise guidance in the form of process feedback can lower ambiguity and assist competitors to narrow the search space, while outcome feedback expands the search space because it leaves plenty of room for interpretation.

Very late in the contest, though, the positive relationship between process feedback and submission affinity disappeared, and may have even flipped to the negative; the professors speculate this may be due to a demotivating, “now-you-tell-me” effect.

Shifting gears from quality to quantity, Sanyal and Ye discovered that both process and outcome feedback encouraged more submissions on the whole. However, they did so in different ways.

Process feedback lured new contributors to the contest; outcome feedback spurred more submissions per contributor. But, again, both of these effects were weakened when feedback was offered late in the game. Interestingly, this contradicts previous studies, which suggest early feedback discourages new contributors from joining. Shun and Ye point out that those studies used only numeric feedback. “We show that when it comes to textual feedback, it should be provided early in the game,” Ye says.

He also comments, “What we find here can very well apply to a traditional context where, say, in an organizational setting, a manager wants a creative solution, or holds a brainstorming session.

“If managers feel that the submissions are converging very quickly, but they want more innovative solutions, they can provide outcome feedback. Or they may observe, ‘Wow, the submissions are all over the place. Doesn’t look like it’s close to what I have in mind.’ Then it’s best to start to provide some process feedback.”

Whichever feedback type they choose, managers should offer it promptly so as to maximize the impact. At the same time, they should be careful to avoid turning their preferences into self-fulfilling prophecies through strongly worded process feedback.

Sanyal uses an illustrative example from his own life: “Many times, if my kids are stuck with something, I hear them and I say, ‘You are on the right track. I won’t tell you the solution, I will only tell you that you’re on the right track.’ So give some overall ideas, but don’t constrain the solution space too much.”

Their work was published in Information Systems Research.

Topics

More School of Business Faculty Research

Are U.S. ‘news deserts’ hothouses of corruption?
November 26, 2024
When CEOs are haunted by memories of past recessions
November 19, 2024
MS in Finance students answer burning investment questions
October 22, 2024
The work-from-home blues have a secret source: nostalgia
September 19, 2024
When expressing gratitude, it’s all in the timing
September 4, 2024

In This Story

Pallab Sanyal

Shun Ye

No easy answers for retailers’ returns dilemma

Jennifer Anzaldi — Tue, 29 Nov 2022 14:48:38 +0000

No easy answers for retailers’ returns dilemma Jennifer Anzaldi Tue, 11/29/2022 - 09:48

Mehmet Altug, an associate professor of operations management, has been researching retail returns policies for a decade. The issue has recently come to prominence, as the lenient policies of online retailers have led to skyrocketing return rates (now exceeding 20 percent in the U.S.). Altug’s various academic papers delve into the difficult trade-offs retailers face when setting returns policies. While there are no easy answers, Altug’s research identifies factors that can help retailers achieve more strategic flexibility.

Mehmet Altug

Who’s smarter—online shoppers or their counterparts in physical stores? On the one hand, e-commerce consumers have a huge advantage in comparison shopping. They can cross-reference products, learn about competing brands and scout for deals in minutes from their living room couch. However, basic physical features of a product—its size, texture, fit, etc.—aren’t represented well on digital devices. That’s why purchasing items such as clothing online can be a stab in the dark.

Retailers have used lenient return policies to counteract e-commerce’s sensory-deprivation problem. Such policies effectively create an antidote for buyer’s remorse, so that consumers can confidently click “Buy Now” for an item they’ve never seen or used. But the industry’s permissiveness regarding returns may be backfiring.

According to the National Retail Foundation (NRF), the average rate of return for online purchases in 2021 was 20.8%, up from 18.1% the year before. The total value of returned items for 2021, according to retailers’ estimates, was more than $761 billion.

Mehmet Altug, an associate professor of operations management at ��AV's School of Business who has been researching pricing and retail operations—and consumer returns in particular—for over a decade, says, “When I first started doing research on this, I used to say up front ‘pricing is the main decision, returns are just something to deal with.’ But now, it’s becoming a primary problem for retailers.”

The main reason why return management does not have a well-defined solution is that it can be seen as a "double-edged sword." While flexible consumer return policies increase our valuation and willingness-to-pay for the product (a positive effect), they also lead to more returned units, which are ultimately sold by retailers at a loss (negative effect).

Therefore, we see a wide range of return policies in the industry. Recently, more and more brands have been tightening up their return policies. For example, Banana Republic, J. Crew, Old Navy and Gap have shortened their grace period to one month; in the U.K., Zara recently started charging for items returned through third-party drop-off points. But it remains to be seen whether online customers, with their penchant for comparison shopping, will accept this austerity drive, or flock to competitors with more permissive policies.

As Altug’s published research shows, the strategic buying behaviors of online customers further complicate retailers’ returns dilemma. His 2016 paper in Manufacturing & Service Operations Management (co-authored by Tolga Aydinliyim of Baruch College) details how offering lenient refunds can entice buyers to purchase items earlier, instead of waiting for them to go on sale. Forward-thinking consumers, however, will anticipate that liberal return policies will lead to more items being returned and marked down for resale. This expectation would increase wait-and-see behavior, as consumers monitor for products to become available at clearance prices. In short, lenient return policies can (again) cut both ways in the presence of strategic customers, either raising or lowering profitability and depending largely on the salvage prices resold items can command through off-price outlets such as TJ Maxx.

These outlets are where department stores turn to offload their unsold inventory, including returned items. And like department stores, they range from low-end (e.g., TJ Maxx and Marshalls) to upscale (e.g. affiliated discount branches such as Saks on Fifth and Nordstrom Rack). Altug’s paper argues that the refund offered for a product is closely related to the resale value of the item, and the partial refund offered should not generally exceed that. The best case for retailers would be a no-returns policy on expensive items with low resale values, since these items would be most subject to the wait-and-see effect among strategic online consumers.

However, retailers are rightfully cautious about the customer defections that may result from restrictive returns policies. Altug’s research finds several alternatives they could explore to preserve profitability in the face of the returns dilemma. For example, if the retailer can enjoy profitable clearance sales (i.e. resale value is greater than the cost of the product) or if the lenient return policies stimulate additional demand, then the retailer may offer full refunds. Similarly, to sustain a lenient return policy, the retailers can lower transaction costs for the consumer by providing return labels in packages. Moreover, in the presence of competition, Altug finds that the retailers’ ability to clear inventory at a higher price (maybe through a reputable clearance partner) will also help them offer higher refunds, stock and sell more at a higher price and make more profits compared to their competitors who do not have the same higher resale value advantage.

The 2016 paper assumes that consumers are making choices in good faith. But free and easy returns also invite opportunists whose intent is to use online retailers as a de facto rental service. Though this class of bad actors is relatively small–responsible for about 11 percent of returns nationwide, according to NRF estimates–it can do significant damage within the low-margin retail industry.

Altug’s 2021 paper in Management Science (co-authored by Tolga Aydinliyim and Aditya Jain of Baruch College) tests the respective profitability of two approaches retailers employ to manage opportunistic returns. First, some retailers use data analytics to identify renters based on their buying and return behaviour and target them through more restrictive returns policies. While mostly effective, this solution carries a risk of false positives, which could unfairly target valued customers. Making inferences about consumer intent from purchase and return information alone can be a dicey prospect. Second, online retailers can deploy price/refund pairing as a sorting mechanism, charging less for an item (e.g., a three-percent discount)–if the buyer agrees to waive free returns. That way, renters self-select to pay up front for their intended temporary use of the product and the honest majority are rewarded, at least initially, with a lower price.

Again, resale price surfaced as a pivotal factor in Altug’s findings alongside renters’ valuation. The self-selection approach works just as well or better than data-driven targeting in a market with low-rental-value potential or when the retailer sells returned units at a low resale price even in a medium-rental-revenue-potential market such as electronics or fashion products.

While there are no easy answers for retailers, Altug’s research maps out the main elements they should consider when setting their returns policy. “There are several issues that need to be considered in this context and still not very well-defined solutions that could address all of them,” Altug says. “One thing is for sure, you can’t just give maximum flexibility and forget it. Retailers need to find the right balance and that is not a trivial task.”

Topics

Tech Talent Investment Pipeline (TTIP)

More School of Business Faculty Research

Are U.S. ‘news deserts’ hothouses of corruption?
November 26, 2024
When CEOs are haunted by memories of past recessions
November 19, 2024
MS in Finance students answer burning investment questions
October 22, 2024
The work-from-home blues have a secret source: nostalgia
September 19, 2024
When expressing gratitude, it’s all in the timing
September 4, 2024

In This Story

Mehmet Altug

Building a 'digital twin' of the real world

Jennifer Anzaldi — Mon, 26 Sep 2022 14:02:39 +0000

Building a 'digital twin' of the real world Jennifer Anzaldi Mon, 09/26/2022 - 10:02

In This Story

Jingyuan Yang

As artificial intelligence (AI) has become more adept at anticipating people’s wants and needs, predictive algorithms have been mingled with almost everything we do digitally. When we are done watching a Netflix show, they take the liberty of queueing up the next one—they finish our sentences in Outlook and Gmail.

Jingyuan Yang

In the offline world, however, the complex and multidimensional nature of many of our most pivotal decisions defies algorithmic analysis. That is, unless AI can learn to detect how real-world contingencies, such as specifics of time and place, govern our choices.

Jingyuan Yang, an assistant professor of information systems and operations management at ��AV School of Business, is at the forefront of AI research that aims to crack the codes of the physical world. Her results so far point toward innovative solutions for some of the biggest societal, governmental, and business challenges we face.

Several of her papers to date investigate urban bike sharing, a “last mile” extension of public transport systems designed to coax commuters out of their cars. Some early adopters of bike sharing, such as New York City and Taiwan, have seen long-term success with the model. But elsewhere, including major Chinese cities, oversaturation has led to bicycle-flooded sidewalks or, even worse, rivers and vacant lots turning into bicycle graveyards—an environmental disaster that produced friction with local politicians.

Yang discovered that in Shanghai, part of the problem was that the distribution of bikes across the city did not match demand. Some areas had far more available bikes than riders, while in others the opposite was the case. Additionally, Shanghai’s system allowed commuters to dismount wherever they chose. This dockless model made predicting rider demand even more challenging, as bikes could be located virtually anywhere.

With a team of six other researchers[1], Yang developed a data-driven model for predicting traffic flows within dockless bike-sharing systems, based on a dataset provided by leading provider Mobike. Spanning the period February 2017-March 2018, the Mobike data contained more than 957 million riding records from nearly 315,000 shared bikes.

The research team extracted flow patterns from the data by partitioning the city and “smoothing out” areas with the lowest levels of activity. The resulting grid-like “flow matrix” carved Shanghai’s bike traffic into spatio-temporal snapshots that could be studied and compared. After clustering these based on their similarities, the researchers could construct “base matrices” that provided broader, deeper points of reference for prediction than temporal or geographic cues alone. Using the base matrix, the algorithm could identify emergent patterns in a certain area as those associated with central business districts on a rainy holiday morning and forecast bike traffic in that area over the next few hours accordingly.

Yang says she was surprised by what this technique of algorithmic mapping revealed. “There are a number of surprising factors we can discover that cannot be covered by traditional model analysis. As an example, we find that there is a slight increase in bike traffic near subway stations during rain, because people want a shorter commute,” she says.

The team tested their model against six other algorithms designed for the same purpose, to see which was the most accurate at predicting actual bike traffic flow in Shanghai. Yang’s team’s solution consistently outperformed the rest on sample datasets for regular working days, rainy working days and holidays–meaning it achieved the lowest degree of prediction error. Perfect predictions are impossible, because all sorts of irregular real-world occurrences, from auto accidents to one-off public events, can cause traffic on a given day to break with the pattern. “Really odd events, we cannot capture,” Yang says. “But the base matrix lets us capture basic trends.”

For Yang, optimizing bike sharing is part of a necessary push toward environmentally sustainable options for urban living, including fewer polluting modes of transport. “All these papers are intended to help companies go a more sustainable way and help the user to tackle the last mile in an eco-friendly manner-without waste and damage to the environment.”

With some tweaks to the logic, however, the same AI-based methods can apply to a range of pressing business issues. The territory mapped by algorithms need not be geographical; researchers can also “map” a network of individuals or companies. Yang’s experiment in the field of B2B marketing is a case in point. She helped build[2] an automatic recommendation engine for marketing campaigns based on customer profiles (similar in concept to the bike-sharing base matrices) reflecting corporate affiliation as well as individual employee status. Customers from the same “region” on the grid, i.e. the same company, are treated holistically to improve recommendation quality.

“When you’re selling to a company, you’re usually dealing with a group of decision-makers who are at different buying stages," Yang explains. "To estimate buying propensity, you need to consider that they may share information. Their behavior should be considered together. Therefore, we use matrix representation to extract their shared knowledge.”

One of Yang’s current research projects focuses on predicting employee flow within networks of companies, again borrowing spatio-temporal techniques. Similar to Shanghai’s urban environment, the job-hopping professional grid has its own version of “weather”—favorable or gloomy economic conditions – that may alter the pattern.

“Based on different job positions, you can group the companies. You can aggregate company profiles and predict, collectively, how many people will leave based on the stock price,” she says.

Yang’s research suggests that by building a chessboard-like “digital twin” of the real world, spatio-temporal AI solutions can help business and society predict–and thus prevent–harmful losses such as human capital flight and damage to the natural environment.

[1] Jingjing Gu, Qiang Zhou and Yanchao Zhao (of Nanjing University of Aeronautics and Astronautics), Yanchi Lui and Hui Xiong (of Rutgers University), Fuzhen Zhuang (of Chinese Academy of Sciences)

[2] In collaboration with Chuanren Liu (of Drexel University), Mingfai Teng and Hui Xiong (of Rutgers University) and Ji Chen (Google).

Topics

TTIP

Tech Talent Investment Program

Impact Fall 2023

At the frontiers of public health

Jennifer Anzaldi — Tue, 16 Aug 2022 11:54:21 +0000

At the frontiers of public health Jennifer Anzaldi Tue, 08/16/2022 - 07:54

In This Story

Ajay Vinzé

Dean Ajay Vinzé

As we all now know from bitter experience, there’s more to proper pandemic response than pills and vaccines. Medical authorities must identify potential hot spots before widespread infection becomes manifest, through early-detection methods such as wastewater testing. With these warning signals, health departments can take preventive action – lockdowns, school closures, etc. – to stop the spread. To build the systems that can make this possible, skilled epidemiologists need to work side-by-side with data scientists who can help them crunch the numbers and predict possible outcomes of various interventions.

Long before COVID was a household word, Dr. Ajay Vinzé, now dean of Mason’s business school, helped pioneer just such a collaboration with public-health officials in Maricopa County, which includes Phoenix and is the fourth-largest county in the United States by population. Vinzé calls this nearly decade-long partnership “a major part of my research and professional journey.”

According to Vinzé, the collaboration began in late 2002 with a missed flight in the Dallas airport. He found himself sitting next to another Phoenix-bound traveler, and the two struck up a conversation. At the time, Vinzé was an associate professor at W.P. Carey School of Business at Arizona State University researching artificial intelligence applications for business problem-solving, including how viruses spread on computer networks, in close collaboration with companies such as Intel and IBM. One of his ongoing research projects focused on failure analysis and failure identification (FA/FI). As Vinzé described his work, his impromptu travel companion – who, as it turned out, was Jonathan Weisbuch, then Director and chief health officer for the Maricopa Department of Public Health – listened closely, asked pertinent questions, and finally invited Vinzé to sit down with a group of leading Arizona epidemiologists.

Why would a top health officer for Arizona’s largest county want his colleagues to listen to an information systems professor? Because Weisbuc recognized that, as Vinzé says, “Computer networks are similar to human systems. Failure means not being able to detect early on that certain propagation is imminent. That’s the same thing you do in computer security.” The two fields also share basic preventive principles and best practices. “To inoculate yourself against viruses on the computer, you need to know the characteristics of the virus, how it spreads, and what happens after infection … and pertinent policy to control the same. Same thing for humans!”

Sarah Santana, Maricopa’s director of epidemiology, found common cause with Vinzé, enlisting him and his colleagues in her department’s efforts to refine post-9/11 surveillance protocols for handling the threat of bio-terrorism. Their work together ultimately led to an article in Information Systems Frontiers (co-authored by Minu Ipe and T.S. Raghu) examining the parallels between “information supply chains”, e.g. Arizona’s state-wide MEDSIS online public-health surveillance network, and e-commerce platforms such as Amazon. In both cases, so-called “information intermediaries” can play a vital role including, but not limited to, aligning incentives and reducing information asymmetry among multiple stakeholders. The paper details how Arizona’s public health department acted as intermediary, paving the way for Maricopa County to join the state-wide system instead of managing its own data independently.

Vinzé gives Santana immense credit for her foresight and openness in allowing ideas imported from business research into public-health emergency preparedness and response. In turn, Santana introduced Vinzé to contexts and questions from beyond academia that widened the aperture of his research lens.

The particularities of the human organism, after all, rarely enter into information systems research but are pivotal to the success of public-health initiatives. Vinzé observes that this comes into play when deciding on key metrics: “How do you attribute value to saving one human life?” The human element also brings a host of contingencies and complications that aren’t salient when dealing with machines.

Vinzé’s gradual coming-to-grips with these intricacies culminated in a 2012 paper for ACM Transactions on Management Information Systems (co-authored by Hina Arora and T.S. Raghu) modeling dynamic interactions between disease transmission, pharmaceutical interventions (e.g. vaccines), and policy-based interventions (e.g. school closures). “It’s a rich simulation exercise – we built the system and turned it over to the county,” Vinzé explains. Using actual characteristics from the 2009 H1N1 pandemic as a test case for various policy approaches, the mathematical model generated numerous insights for guiding public-health policymakers not just in Maricopa and the southwest United States, but perhaps worldwide. For example, the simulation showed that rolling school closures of two to four weeks were not only less effective at reducing transmission than county-wide, eight-week closures – but they also resulted in more weeks of shutdown over the length of the pandemic. Additionally, the results reinforced the researchers’ intuition that the quality of pandemic response hinges upon finding the right mix of pharmaceutical and non-pharmaceutical interventions. Years later, these data-driven principles would be put to the ultimate test during the COVID pandemic.

Over the course of his long collaboration with Santana and the Maricopa health authorities, Vinzé got the chance to view public-health challenges from virtually every angle. “We did survey research for them, sometimes experimental research, sometimes systems-building,” he says. He even served on a healthcare-focused working group as part of the planning process for Super Bowl XLII (2008), and subsequently Super Bowl XLIX (2015) both held in Glendale, Arizona. In 2012, Vinzé was recognized with a Faculty Achievement Award from Arizona State University for “defining edge research and creative work for professional application” recognizing high impact research over a ten-year period – his association with public health was acknowledged as a main driver of impact.

In hindsight, Vinzé suggests that his foray into public health spotlights and clarifies certain aspects of digital transformation. He points out that technological advancement comes in waves, producing sudden surges in computing power, processing speed and data accessibility. With each wave, public- and private-sector actors alike experience a tidal force of opportunity (for leaders) or adversity (for laggards). Collaborations between academia, business and government can help translate technological evolution into positive outcomes for the economy, communities and the wider society.

“Each wave has an implication on the workforce of that time. We’re going through this big wave as we speak, coming out of COVID. The question is, are we ready?” Vinzé says.

Topics