Professors Honored for Research and Teaching Excellence

Though he excels in a discipline notable for its focus on numbers, Rani Hoitash, a member of Bentley’s Accounting department since 2007, has distinguished himself by prioritizing people. “As a social science researcher, I’m most passionate about how the human element influences outcomes in corporate governance, audit regulation and capital markets,” he says.  

Among the 40-plus peer-reviewed articles Hoitash has published in top-tier academic journals, his most cited paper explores how intensity of corporate board monitoring affects a firm’s value and other outcomes. He has also examined how employees’ trust in management affects the quality of a firm’s financial reporting, the association between auditors’ health and auditing outcomes, and how gender and racial identity influence auditors’ career advancement. His findings have been widely reported in in mainstream media outlets, including The Wall Street Journal and Forbes, and cited in official reports from the Securities and Exchange Commission, Public Company Accounting Oversight Board and other U.S. regulators.  

Colleagues and students alike praise Hoitash for his unwavering support of their own research efforts. A former chair of Bentley’s Research Council, he helped implement new funding opportunities for faculty, distributing hundreds of thousands of dollars in support for new projects. Hoitash has also directly supervised or served on dissertation committees for eight graduates of the university’s PhD in Accounting program — many of whom he counts among his research partners today.

One of those graduates is Jenna Burke PhD ’17, a tenure-track assistant professor at the University of Colorado Denver. “His expertise in auditing and continued mentorship have immeasurably contributed to my success as a scholar,” she says of Hoitash, with whom she’s co-authored several papers. “I have benefited from observing his dedication to research and, on a more personal level, from our discussions about career aspirations, work-life balance, and navigating the intricacies of academia.”  

Following a distinguished career as an educator and administrator — most notably, as headmaster of Phillips Academy in Andover, Massachusetts, and president of Guilford College in Greensboro, North Carolina — Don McNemar joined Bentley’s Global Studies department in 2002.  

A devout Quaker, his religious beliefs were fundamental in fostering his interest in international diplomacy. “Quakers have a basic belief that God is in everyone and that everyone should be treated with respect,” McNemar once shared. “We also believe when cultures are different from your own, you must try to understand that culture and want to bring peace between nations.” He emphasized the importance of cross-cultural understanding through the classes he taught and as co-adviser of Bentley’s Model U.N. chapter.  

“Don brought an experiential, real-world approach to all of his courses, helping students develop critical skills in teamwork, negotiation and public communication while highlighting global issues of ethics, cross-national justice and inequality,” says colleague Joao Resende-Santos, an associate professor of Global Studies who served as McNemar’s Model U.N. co-adviser for nearly 20 years. “Few other Bentley faculty have been as devoted our students and as effective in nurturing their learning.” 

Stakeholder support, a prerequisite for the long-term success of any business, is particularly important for social enterprises (SEs), says Jill Brown, director of Bentley’s Executive PhD in Business program.  

In “The Fork in the Road for Social Enterprises: Leveraging Moral Imagination for Long-Term Stakeholder Support”— which appeared in the journal Entrepreneurship Theory and Practice and received the International Association of Business and Society’s 2022 Best Published Paper Award — Brown and her co-authors take a closer look at the complex and nuanced stakeholder networks of SEs and propose a conceptual process model to ensure long-term success. Their model, which distinguishes between primary and secondary stakeholders, requires social entrepreneurs to use moral imagination, or the ability to evaluate the potential helpful or harmful effects of business activities from multiple perspectives.  

“Many SE ventures, once they become operational and begin having real-world impact, end up having unintended effects the founder does not anticipate,” Brown notes. Thus, SEs “must continually formulate new plans and ways of collaborating that enable them to uphold their core values and continue to meet the expectations of both primary and secondary stakeholders.”  

Long before ChatGPT became a permanent fixture in media headlines, Noah Giansiracusa was questioning the ethical implications of companies using AI algorithms that promulgate “fake news” — fabricated or misleading information that masquerades as fact.  

In his 2021 book, “How Algorithms Create and Prevent Fake News: Exploring the Impacts of Social Media, Deepfakes, GPT-3, and More,” he explains how the rise of digital media and advances in machine learning techniques have eroded the average American’s ability to separate fact from fiction, increasing both the speed at which fake news spreads and the magnitude of its influence. Every time we visit social media sites like Facebook and YouTube, Giansiracusa says, we leave behind a “trail of digital crumbs” based on what we like, share, read or watch on each platform; AI algorithms use this information to decide which ads or posts we see in our Facebook news feeds and which videos appear in our YouTube recommendation queues, a pathway that has allowed fake news to flourish.

Giansiracusa believes federal oversight is necessary to ensure that companies become more transparent with their algorithms — and accountable for their actions. Since publishing his book, Giansiracusa has prominently and publicly advocated for AI regulation. He’s authored dozens of op-eds in leading media outlets, including TIME, The Washington Post, The Boston Globe and Scientific American, and was invited to share his expertise with international policymakers during an Organization for Economic Cooperation and Development (OECD) conference last fall.

“Correlation does not imply causation” is the Golden Rule of statistics. This simple maxim reminds researchers working with complex data sets to carefully consider the relationship between two variables and rule out the existence of a confounding factor, or confounder: a third, unanticipated variable that affects the other two — and distorts the true nature of the relationship.    

Known as causal inference, the process of establishing a genuine cause-and-effect relationship between variables is made exponentially more challenging when the data sets in question are text documents, says Reagan Mozer; the inherent complexity and subjectivity of language make it difficult to identify specific attributes suitable for direct comparison.

In “Matching with Text Data: An Experimental Evaluation of Methods for Matching Documents and of Measuring Match Quality” — which appeared in the journal Political Analysis and received the Society for Political Methodology’s 2021 Miller Prize for best published paper — Mozer and her co-authors conduct systematic evaluations of more than 100 unique text-matching methods and provide a general framework for optimizing results. Through two separate evaluation studies, one measuring partisan bias in mainstream media outlets and the other measuring outcomes of a medical intervention, the researchers demonstrate the utility of text matching through a combination of text representations and distance metrics. 

Recent studies indicate that adolescents and young adults who use social media are at increased risk for suicidal behavior. But many researchers, including Bentley’s Jennifer Xu, believe online platforms can be harnessed to help users experiencing emotional distress. 

In “Finding People with Emotional Distress in Online Social Media: A Design Combing Machine Learning and Rule-Based Classification,” published in the journal MIS Quarterly, Xu and her co-authors from the University of Hong Kong introduce a comprehensive classification system to help mental health professionals in China identify at-risk users more efficiently and effectively. Their algorithmic model, KAREN (Karen Automated Rating of Emotional Negatively), combines machine learning and rule-based classifiers to quantitatively and qualitatively analyze user-generated blog posts written in Chinese.

Using KAREN, researchers were able to correctly identify 40% more individuals experiencing emotional distress than traditional search engines — a testament to the program’s life-saving potential. As Xu notes, “Our proposed approach makes it possible for suicide prevention programs to provide more proactive and effective services and prevent tragic incidents from happening.”