Ethical mathematics awareness in students’ big

data decision making

Michelle Stephan, University of North Carolina at Charlotte,  mstepha1@uncc.edu
Jordan Register, University of North Carolina at Charlotte
Luke Reinke, University of North Carolina at Charlotte
David Pugalee, University of North Carolina at Charlotte
Lenora Crabtree, University of North Carolina at Charlotte
Christine Robinson, University of North Carolina at Charlotte
Premkumar Pugalenthi, Palisades Episcopal Schools

In this paper, we investigate adolescent students’ ethical mathematics awareness as they reason
through interview tasks that provoke them to make a decision based upon the results of data
analytics. Drawing on multiple resources, including ethical principles from numerous STEM and
business disciplines, we introduce an analytic framework for documenting the ethical
mathematics awareness of adolescent students. Our findings indicate that 1) students justify
their decision using a variety of ethical principles, and 2) half the students’ decisions benefitted
their hypothetical employer and half protected society. We conclude the paper with a discussion
of implications for designing future instruction to support students’ growth in ethical
mathematics awareness.

Scholars in mathematics education maintain mathematics is “well integrated into the techno-
logical, industrial, military, economic and political systems of the present ‘Westernized’
world” (D’Ambrosio, 1998, p. 67). According to Lengnink (2005) the systemic influence of
mathematics is a result of its embeddedness in the technology that has come to shape our
world by unconsciously affecting our thinking and behavior. Unfortunately, the assumed
objectivity of mathematics often generates “dehumanized thinking” which lends itself to
“instrumentalism […] ethics-free governance and social practices” (Ernest, 2018, p. 205).
Benjamin (2019) appropriately labels this phenomenon The New Jim Code, referring to the
use of technologies and, by default, mathematics to reproduce existing inequities. As such,
the effects of mathematics and technology on human behavior and well-being require that
mathematics become a morally and ethically grounded discipline.
Literature on the detrimental effects of mathematics on society has been expanding since
the onset of globalization (D’Ignazio et al., 2020; Wheelan, 2014). O’Neil (2016) describes the
discriminatory algorithms and models that have come to organize, manage, and make
Please cite as: Stephan, M., Register, J., Reinke, L., Pugalee, D., Crabtree, L., Robinson, C., & Pugalenthi, P.
(2021). Ethical mathematics awareness in students’ big data decision making. In D. Kollosche (Ed.),
Exploring new ways to connect: Proceedings of the Eleventh International Mathematics Education and
Society Conference (Vol. 3, pp. 977–985). Tredition. https://doi.org/10.5281/zenodo.5416678
M. Stephan et al.

decisions about human lives. These mathematical models, intended to increase productivity
and reduce human error, have replaced human decision-making processes in areas such as
loan disbursement, insurance qualification, policing, and political campaigns. Because such
algorithms are built using statistical correlations as proxies for desired information, groups
who have been historically targeted for their perceived socially unacceptable behavior
become trapped in a vicious cycle of self-fulfilling prophecy. In 1973, Paulo Freire warned
that the influx of new technologies had the potential to develop massified consciousness, or
an individual’s flawed belief that humans act through free choice alone rather than by choice
and manipulation. Individuals with massified consciousness participate in their own
domination by uncritically accepting media reports rather than through their critical
reflection on the world. To overcome such thinking, Freire (1970) introduced a pedagogy
aimed at developing, critical consciousness, whereby an individual learns to “perceive social,
political and economic contradictions and to take action against the oppressive elements of
reality” (Freire, 1970, p. 36). Mason (1986) claims that it is the responsibility of educators to
ensure that citizens possess the “intellectual skills to deal with information” including the
ability to read, write, reason, and calculate (p. 10). Given the intimate role that mathematics
plays in politics and policy, civic life, Information and Systems Technology, as well as the
number of mathematics majors who go on to work in related industries, we argue that
mathematics education should play a considerable role in preparing students for their ethical
responsibilities as well.
The majority of the literature on ethics in mathematics is theoretical and/or philo-
sophical. Few empirical studies exist which incorporate ethics into mathematics learning;
the majority of these exist in the realm of Critical/Social Justice mathematics (Frankenstein,
1983; Gutstein, 2006; Kokka, 2020; Rubel, 2017). In this paper, we explore the ethical
mathematics awareness of 13 middle and high school students of racial and economic
privilege by analyzing their responses to tasks that involve making decisions with the results
of data analytics. For this exploration, we developed an Ethical Reasoning in Mathematics
(ERiM) analytic framework that is specific to data science. In this paper, we present the ERiM
framework and use it to analyze adolescents’ ethical mathematics reasoning.

Methodology
Our theoretical perspective is situated within the Critical Mathematics education (CM)
program elaborated by Frankenstein (1983) and Shor (1993). The complementary goals of
CM are to a) promote students’ capacity for critical reflection so that they can question
mathematical arguments, models and representations and b) inspire students’ agency to act
in ways that use mathematical communication to liberate, rather than disenfranchise,
groups. We define the term critical mathematics consciousness (CMC) as a content-specific
form of critical consciousness (Freire, 1970; 1973) that involves the awareness of the role that
mathematics plays in disenfranchising or liberating oppressed groups in society and the
willingness and commitment to act. CMC involves three different types of awareness
(Stephan et al., 2021):

978
 Ethical mathematics awareness in students’ big data decision making

− Sociopolitical Mathematics Awareness that mathematics is used to model and interpret

the real world and can be used to make decisions both at the individual and systemic
levels that may be oppressive or liberatory.
− Ethical Mathematics Awareness that human beings do mathematics; thus there are
potential ethical dilemmas and implications of mathematical work.
− Communicative Mathematics Awareness that mathematical communication has the
power to educate and mis-educate society and encourage the masses to act in certain
ways.
Like Freire, we view critical mathematics consciousness as something that can be
developed with critical pedagogies.

Analytic framework
Ethical mathematics awareness refers to being aware of ethical implications that may or may
not be considered in the process of making mathematical decisions, such as violations of
privacy, questions of who owns the mathematical data, and bias in the data, to name a few.
To create the Ethical Principles Framework in Table 1, we consulted a variety of resources
including philosophy and mathematics education articles. While the general principles
expressed in these resources were useful, we also suspected that more nuanced principles
pertaining to the STEM professions might be found in materials published by professional
organizations for engineering, data analytics, artificial intelligence, computing and
information, science, and statistics. Since many of our interview questions employed a
business scenario as the context for making decisions, we also explored resources from
business domains including marketing, accounting, and finance.
The Ethical Principles Framework does not represent a finite list of principles, but rather
a compilation of the most relevant principles for ethical reasoning in the intersection of
STEM and business professions. One point of elaboration involves the distinction between
fairness and discrimination. Fairness involves attempts to avoid biased decisions and thus
unequal treatment towards human beings. Discrimination refers specifically to the
unfairness committed upon already-oppressed groups in society.

Participants
The work presented in this article was conducted as the first step in a Design Research
Project (Cobb et al., 2003) that focuses on Designing for Critical Mathematics Consciousness
(CMC) among students of privilege. Interviews were conducted primarily with 14–16-year-
old students with economic and racial privilege to a) understand their CMC and b) determine
viable contextual problems that may provoke critical mathematics consciousness. The
Design Research Team consists of two mathematics educators, two STEM educators, one
doctoral student and one assessment professional. We interviewed thirteen students, nine of
whom identified as male and four as female. Nine of the students matriculated at Hill High

979
M. Stephan et al.

Charter 1 high school, three from a Lakeview Charter middle school and one from Paradise
Bluffs, a private, middle school. Thirty percent of Hill High Charter’s students perform at
grade level in mathematics and the student population is 65% White, 22% Black and 5%
Hispanic. Sixty-three percent of Lakeview Charter’s students perform at grade level in
mathematics and the student population is 70% White, 13% Black, and 5% Hispanic. Finally,
eighty percent of Paradise Bluffs’ students perform at grade level in mathematics and the
student population is 75% White, 13% Black, and 1% Hispanic.

Professional Domains
Ethical Principles Considerations
Addressing the Principle
Privacy How does respect for freedom or personal autonomy NSPE*, JMU
apply? Is confidentiality required? Is consent needed ASA, ACM/AI
or obtained? Is privacy violated?
Fairness What is the fair or just thing to do? Is there fair ACM, AIB,
(equality) access to systems that are created? Data Analytics, JMU
Accuracy Is data reliable and accurate? NSPE, AIB,
ACM/AI, Science,
Accountability Who is accountable? Have they communicated the Data Analytics
data in a misleading manner? Is the source reliable?

Property Whose data is it to sell? Who owns the data? Data Analytics
Loyalty Is the decision/activity loyal to the organization (e.g., AIB, JMU
makes profit, keeps ideas within organization, does
not use organization’s ideas to make money
elsewhere)
Accessibility What information does an organization have the Data Analytics
right to access about people? Who has access to this
data? User/buyer?
Algorithm Bias Are algorithms objective? Do algorithms AI
(un)knowingly discriminate against individuals or
groups?
Transparency Are the codes for algorithms readily available for ACM/AI, Science, ASA
inspection?
Ecological Has the impact on humans and ecosystems been ACM/AI, NSPE
considered?
Employment Will the decision/activity harm an individual’s or AI
group’s employment status?
Discrimination Has the decision/activity avoided negative effects on Data Analytics, ACM/AI,
oppressed societal groups? AIB
*James Madison University Ethical Reasoning in Action (JMU); American of Statistics Association (ASA); Resnik
(1998)-Science; Chessell (2014)-Data Analytics; National Society for Professional Engineers (NSPE); Association
for Computing Machinery (ACM); Artificial Intelligence (AI); International Business (AIB).

Table 1. Ethical Mathematics Reasoning Framework

1 All student and school names are pseudonyms.

980
 Ethical mathematics awareness in students’ big data decision making

Method
Interviews lasted from 25-45 minutes and took place at the student’s school in a private
setting during a non-academic class period such as physical education. The interviewer
began each session with general talk, attempting to build rapport with the student, and then
posed five tasks. Due to the brevity of the paper, we only elaborate on students’ ethical
awareness on the Great Groceries task below.
Great Groceries is one of the biggest successes in American grocery chains. Impressively, Great Groceries
reported $14.74 million in quarterly profits last year. You report to the CEO for Great Groceries. One of
his goals for 2020 is to increase profits.
Your Management Team, based on extensive data analysis, comes up with the following three
recommendations for increasing profits:

Option A: Tiered Membership Fee to shop at the store. Offer Option B: Raise the price on
tiered membership fees, the idea being that richer customers the five top selling items by
buy more products. The Tiered-Membership Program would $1.50 each, increasing profits
increase profits by 1.8%. by 0.5%.

Member’s Yearly Membership Option C: Put more digital,

Income Level Fee per Year self-serve checkout stands in
Upper Tier $150,000 and up free their stores to decrease the
Middle Tier $70,000-$150,000 $50 number of human workers,
Lower Tier $30,000-$70,000 $75 raising profits by 3.5%.
Sub Tier Under $30,000 $100

Which option would you recommend to your CEO, if any? Explain. Why didn’t you suggest Option A, B,
or C?

The options were designed intentionally to create a potential ethical dilemma for students
between choosing an option that placed profits over people. Option A may prompt students
to recognize issues of economic classism and structural oppression by imposing a shopping
fee on the poorest customers. Option B, in our view, posed the least harm to people but also
raised the least profit. Option C would generate the most profit but also eliminate jobs. When
a student made a decision, the interviewer posed follow up questions to elicit their
justifications. Each interview was transcribed and three team members used the Ethical
Mathematics Reasoning Framework to categorize each student’s responses independently.
Once coding was completed, the three team members met to compare results and achieve
consensus.

Findings
We organize the findings around two main themes. First, there were two main concerns that
informed students’ decisions: Consequences for a) humans and b) profits. Second, there were
a variety of ethical principles invoked by students as they made their decisions.

981
M. Stephan et al.

Impact on Humans or Profit?

Of the thirteen students who answered the Great Groceries question, six picked the option
that they considered to have the least harmful impact on humans. Two students indicated
that they would choose option C because they felt their personal employment status
depended upon recommending the option that would bring in the largest profit; these
students indicated that if they did not have to report to the CEO, they would choose the least
harmful option. The remaining five students chose the option they perceived to be best for
the business, in terms of increase in profits.
Jamal’s reasoning is one example of instances where students anticipated how their
decision might negatively impact humans.
Jamal: I pick option B [be]cause like, um, option C taking people’s jobs didn’t sound
right and because people need money too, to help their families. And option
A, just helping the richer customers is, it’s not really cool.
For Jamal, Option B offered the best possible outcome for people “because it doesn’t seem
like it’s doing anything bad…” However, Option A was not fair because some people, most
notably the rich, were getting something for free while others had to “pay a fine.”
While five students eliminated Option C, seven other students articulated their loyalty to
the company by choosing it. For example, Collin recognized the impact that choosing Option
C would have on humans, yet made his decision based upon what yielded the highest profit.
Collin: It’s kind of hard because you’re also putting so many people [out o] the jobs.
But that would also mean you wouldn’t have to pay them. You wouldn’t have
to pay them at all. So yes, your profit would go up. So I guess, C.
Int: All right, how come you didn’t do option B?
Collin: Because if the top five selling goods are already selling at their price and you
bump it up a dollar 50 maybe those people won’t want to buy it anymore.
For Collin, his commitment was to increase profit for the company, despite recognizing
the negative impact on employment of the workers. He continued to rationalize his decision
by arguing that minimizing the number of workers would decrease salary expenses, raising
profits even more. Even when asked if he did not have a CEO to report to, Collin still chose
Option C.
Jade, on the other hand, changed her decision when given the opportunity not to report
to the CEO. Her initial decision was Option C, possibly due to her belief that her primary
loyalty was to her employer even though it conflicted with her ethical beliefs.
Jade: I think the last option probably would be the best just based on how this one
is 1.8% this is 0.5 and this is 3.5 but at the same time…although this raises
profits it does put a lot of people out of jobs, which also isn’t a good thing for
the rest of everyone…Um, I don’t think that’s [Option B] a bad one, but it
doesn’t have the best increase of profit. And at the same time it does make
things more expensive. So you might lose customers.

982
 Ethical mathematics awareness in students’ big data decision making

Jade: [Interviewer gives her the choice not to report to the CEO] Oh yeah, I would
probably say the first one then because, um, well you’re not forcing everyone
to pay an extra dollar 50 and, if you can afford the membership, then you’ll get
the membership…You don’t have to worry about like the moral part of it. Um
employment, making sure people get paid and can support themselves.
In the exchanges above, Jade recognized the impact that choosing Option C has on the
livelihood of humans due to losing their means of supporting themselves. However, her
loyalty to the CEO and the company compelled her to make an “amoral” decision. Although
she did not recognize the negative impact of Option A on families that live in poverty, she
nevertheless, chose what, to her, was the least hurtful option when the possibility of not
reporting to the CEO was posed.
In summary, about half of the students we interviewed made their decision by
considering the impact that it would make on human beings, while roughly half remained
loyal to the company. Two of those seven however, changed to a less harmful Option when
relieved of the criterion to report to the CEO. As we will see in the next section, students
evoked several ethical principles as they decided which option to recommend.

Other ethical principles

In the previous section, we documented the end result of students’ reasoning on Great
Groceries and learned that the number of students making choices based primarily upon
negative consequences for humans was about equal to the number who considered the
positive effects on the company’s profits. There were six ethical principles evoked by
students as they justified their decision: Employment, Fairness, Loyalty, Accuracy,
Discrimination, and Privacy.
Eight of the thirteen students mentioned the impact that their decision would have on
employment. Five of those eight argued that choosing Option C would negatively affect
people whose means of supporting themselves would be compromised. Another student,
Miya, did not choose Option C, but her reference to employment actually focused on the
negative impact that unemployment has on the economy, not the family. Finally, two
students who chose Option C commented on employment arguing either that people would
lose jobs but profits would go up (Collin above) or did not acknowledge the impact on
workers at all, just increase in profit.
Six students mentioned fairness in their justifications saying that it was not fair that some
people had to pay more than others just to get in the store. Some of the students’ arguments
indicated that they noticed it was the poor who were being made to pay more than the rich
and exhibited empathy for the unfairness. Only one student made connections to systemic
discrimination in this Option which we identify as the Discrimination principle.
Of the six students who used loyalty to justify their decision, not surprisingly, all of them
chose the Option that they perceived would generate the highest profit.
Four students questioned the accuracy of the profit predictions made in some of the
options. Accuracy refers to questioning the data both in terms of reliability and validity.
983
M. Stephan et al.

Although the students did not refer to the data that was used to create these options, they
did question the viability of the options, or as Miya said: “I don’t like Option A. It just doesn’t,
they’re just thinking. They don’t have the math for it. They can’t just say that.” In other
words, Miya suggested that the Option creators were merely conjecturing (thinking) rather
than using data (math). In terms of privacy, only one student mentioned this principle in
regards to Option A. Ella argued that she would choose Option A, over Option C, only if the
customers’ plans could be kept private so that they were not aware of the higher fee they
may be paying (implying a fairness issue).
Only one student made his decision (Option B) based upon recognizing what he perceived
to be discriminatory practices in two of the options.
Tyrone: Definitely not option A…So if even if the idea worked out, you’re cheating out
the lower class who only received $30,000, and they have to pay a membership
fee of $100…that’s a lot of money and no one would really spend that much
money that’s in poverty on their tier with $100. And it’s kinda like taxes,
wealthier, I know a lot of popular companies don’t have to spend money on
taxes, but they’re the ones who have the most. Option B [is] the most fair
route I would go because everything else is just like morally incorrect. That’s
mean, morally incorrect because you’re taxing, um, poverty. And then option
C…I don’t want to decrease jobs because people still need jobs. And, um, I
know a lot of the grocery store occupation is like located with people in
poverty, high-schoolers and retirement, people in retirement. So I don’t think
it’s fair to any of those groups because they’re the ones most troubled…
We categorized Tyrone’s reasoning as recognizing the discrimination that could occur as
the result of carrying out Options A and C. Tyrone attended to the oppression that would be
inflicted (fees; taxes) on already oppressed groups of people (people in poverty) by systems
(wealthy companies; governments). Not only did he foresee the negative impact of his
decision in the Great Groceries Dilemma but also recognized systemic injustices that are
sometimes perpetuated by the taxation system in his country.
In summary, the students employed a variety of ethical principles as they deliberated the
impact of their business decisions. Students almost evenly split on staying loyal to the
company by increasing profits or making decisions that would not impact the livelihood of
citizens negatively. Attending to the employment of citizens was the top ethical principle
evoked by students. It should be noted that two students questioned the ethics of Option B
by considering which particular items are best sellers. For example, Miya and Greta thought
it would be unethical to raise the prices on what were likely top sellers at the time of the
interview (the COVID-19 pandemic), toilet paper and hand sanitizer.

Conclusion
Our analysis indicated that middle and high school students have indeed developed quite a
repertoire of ethical principles and employ them as they consider situations that potentially
call for empathy and subsequent action. At least half of the students chose options that they

984
 Ethical mathematics awareness in students’ big data decision making

perceived would have the least negative consequences for humans with one student noticing
potential systemic discrimination. The students who exhibited loyalty to the company and
its profits either did not notice the negative potential outcomes of the other choices or felt
obligated to choose profits as a part of their job. Or as Tony put it, “when you’re running a
business, it’s about profit.” We see problems such as Great Groceries as rich contexts for
discussions about the accuracy of the data-driven options, issues of fairness and
discrimination and whether profits or the safety and livelihood of citizens should drive data-
based decision making.

References
Benjamin, R. (2019). Race after technology: Abolitionist tools for the new Jim Code. Polity Press.
Chessell, M. (2014). Ethics for big data and analytics. Retrieved on January 8, 2021, from
https://www.ibmbigdatahub.com/whitepaper/ethics-big-data-and-analytics
Cobb, P., Confrey, J., diSessa, A., Lehrer, R., & Schauble, L. (2003). Design experiments in education
research. Educational Researcher, 32(1), 9–13.
D’Ambrosio, U. (1998) Mathematics and peace: Our responsibilities. Zentralblatt ür Didaktik der
Mathematik, 30(3), 67–73. https://doi.org/10.1007/BF02653170
D’Ignazio, C., & Klein, L. F. (2020). Data feminism. The MIT Press.
Ernest, P. (2018). The ethics of mathematics: Is mathematics harmful? In P. Ernest (Ed.), The philosophy
of mathematics education today (pp. 187–216). Springer.
Frankenstein, M. (1983). Critical mathematics education: An application of Paulo Freire’s epistemology.
Journal of Education, 165(4), 315–339.
Freire, P. (1970). Pedagogy of the oppressed. Continuum.
Freire, P. (1973). Education for critical consciousness. Press.
Gutstein, E. (2006). Reading and writing the world with mathematics: Toward a pedagogy for social justice.
Routledge. https://doi.org/10.4324/9780203112946
Lengnink, K. (2005) Reflecting mathematics: An approach to achieve mathematical literacy. Zentralblatt
ür Didaktik der Mathematik, 37(3), 246–249.
Kokka, K. (2020). Social justice pedagogy for whom? Developing privileged students’ critical mathe-
matics consciousness. The Urban Review, 52, 778–803. https://doi.org/10.1007/s11256-020-00578-8
Mason, R. (1986). Ethical issues of the information age. MIS Quarterly, 10(1), 5-12.
O’Neil, C. (2016). Weapons of math destruction: How big data increases inequality and threatens
democracy. Crown.
Resnik, D. (1998). The ethics of science: An introduction. Routledge.
Rubel, L. (2017). Equity-directed instructional practices: Beyond the dominant perspective. Journal of
Urban Mathematics Education, 10(2), 66–105.
Shor, I. (1993). Education is politics: Paulo Freire’s critical pedagogy. In P. Leonard & P. McLaren (Eds.),
Paulo Freire: A critical encounter (p. 25–35). Routledge.
Stephan, M., Register, J., Reinke, L., Robinson, C., Pugalenthi, P., & Pugalee, D. (2021). People use math
as a weapon: Critical mathematics consciousness in the time of COVID-19. Educational Studies in
Mathematics. https://doi.org/10.1007/s10649-021-10062-z
Wheelan, C. J. (2014). Naked statistics stripping the dread from the data. Norton.

985