2. EXPLORING RESEARCH QUESTIONS
We now discuss the responses to the questions in the previous section from the workshop attendees, along with the interpretations of the writing group.
2a. Question 1: What are the core elements of a good module in the context of socially relevant data sets that focus on the African diaspora?
The data and African diaspora modules to be used in the course require a structured sequence that provides students with progressively developing basic data science skills. Data and the African diaspora modules should be intellectually accessible, engaging, and relevant to students' lives. They should focus on real-world issues and have clear motivation. Modules should be self-contained and relate to a common theme, ensuring that goals and learning outcomes are specifically identified. Examples include modules on wrongful convictions or water crises relevant to local demographics. Authentic engagement with real-world issues is crucial, along with the inclusion of culturally relevant datasets.
The flexibility of a module can be harnessed by the incorporation of different courses across the disciplines. We define course modules as sections of the course content, materials, learning activities, and assignments that cover grouped concepts designed to build student knowledge, understanding, and skills. Course module length and complexity may vary depending on the instructor, department, or institution’s preference; however, each module should include three core elements: data science themes, African Diaspora themes, and engaging teaching approaches. The data science themes include data sets, mathematical and statistical techniques, and computer programming. The African diaspora themes include socially relevant content, culturally relevant topics, and economics. To optimize student learning, promising practices for developing and implementing data and the African diaspora module should engage students through intentional pedagogy, course content, instructional design, and relevant data sets. The foundation for the module includes building students' knowledge, beginning with establishing course context and defining relevant terms. The following table summarizes the components of a module.

Data Science Themes
Data Sets: Data type and data structure with variables relevant to the Black community
Math/Statistics: Graphs, points, lines, variables, data wrangling, statistical analysis,
Programming: Reading code, debugging, commenting, coding (e.g., Excel, Python, R/R markdown, Tableau)
Analysis: Visualization, modeling, data curation, ethics, bias, communication, and storytelling
African Diaspora Themes
Social Context: Examining studies on social issues, systemic factors, and justice relevant to the Black community
Cultural Context: Considering variables such as race, ethnicity, geography, demography, and proxy variables that identify race
Economic/Environmental:
Examining the impact of relevant data (i.e., employment, income, and entrepreneurship) on environmental and health justice
Variables: Race, proxy variables that identify race, ethnicity, geography, and demography
Teaching Approaches
Student Engagement: Hands-on interaction with examples and practice
Project-based Learning: Carry out the data science lifecycle to conduct the analysis
Module Framing: Establishing why the focus is on a certain topic and what is considered part of the African diaspora
Balance: Leveraging the African diaspora context to better understand and analyze the data
2.a.i. African Diaspora Content. To provide a context for the African diaspora, participants grappled with a foundational approach to ‘Afrocentricity’ upon which to build the modules. Afrocentricity centers on the cultural and historical experiences of African people and their descendants [2 - 5]. Molefi Kete Asante, a foundational scholar in Afrocentric thought, defines Afrocentricity as "a mode of thought and action in which the centrality of African interests, values, and perspectives predominate." The project leadership team defined “African diaspora” as anyone who originated from the African continent and sought to create a workshop environment that explored addressing data science as it relates to impacts on Black communities.
Unfortunately, data technologies have the potential for racial and other forms of bias, from facial recognition in policing to risk assessment tools in lending [6 - 10]. To drive positive social impact through data, there must be an understanding of how to address historically biased data and develop frameworks for critical course content that prepare students to be aware of the interplay between racism and data. As all disciplines and all fields leverage data, there is an increasing need to educate students to be data-aware and understand how data impacts their world. To steward the educational experiences, there is a need to develop and teach modules that interplay between data science and the African diaspora to actualize social justice for all.

2.a.ii. Data Sets. Data sets relevant to the African American community vary across disciplines and expertise. Often, the data is publicly available through governmental websites and universities or public libraries, while restrictive data sets may require a fee or the use of an IRB for access. In the teaching of the module, there are some key concepts to keep in mind.
Data and data sets are not objective; they answer the questions that are put to them and “are given voice by those who interpret them” [11]. Data sets only exist because someone, some group, and/or some society decided the topic and/or the people and their experiences were important enough to be observed and recorded. To understand data sets and their quality, validity, and reliability, it is necessary to consider the sources of the data, including what data were collected and why, how they were collected, and by whom under what circumstances. It also means understanding what data wasn’t collected. For example, data from cultures with oral traditions are much less likely to be included in data sets, while data of interest to those with the skills and resources to collect and store data are more likely to be included. Data sets are more likely to be designed by those who are part of the dominant culture. Decisions about the type of data to be collected, particularly about Black people and other minoritized people, can be influenced by the deficit model, which is based on the premise that deficiencies in a culture are the cause of differences between minority group members and members of the dominant group [12].
The collection of data itself can be biased. Observers have long been found to rate the same behaviors differently based on the perceived characteristics of those being observed [13], and asking for demographic information at the beginning of a measure can impact participant response [14].
2.a.iii. Socially
Relevant
Versus Culturally
Relevant.
The distinction between "culturally relevant" and "socially relevant" lies in their respective scopes and focuses.
Socially relevant refers to information, initiatives, or actions that address and impact issues or concerns within society. This concept emphasizes the importance of addressing pressing social issues, such as inequality, injustice, discrimination, poverty, environmental degradation, and human rights violations. Socially relevant content or activities aim to bring about positive change, raise awareness, and promote social justice and equity. They may involve advocacy, education, community engagement, policy reform, or other forms of action aimed at addressing societal challenges and improving the well-being of individuals and communities. Socially relevant initiatives
are often grounded in values of empathy, compassion, solidarity, and a commitment to advancing the common good.
Targets broader societal issues, concerns, and challenges that impact individuals and communities across diverse cultural backgrounds. The purpose serves as a catalyst for positive change, raises awareness, and fosters social justice and equity on a societal scale. One example is advocacy campaigns for human rights, initiatives combating poverty and inequality, movements addressing environmental sustainability, and efforts to dismantle systemic racism and discrimination. Socially relevant content refers to educational material, topics, and discussions that address pressing social issues, challenges, and phenomena that impact individuals, communities, and societies.
Culturally relevant content [15 -16] is centered on facilitating students' acceptance and affirmation of their cultural identities while providing critical educational and experiential exposure to the field and can enhance student engagement and persistence. Culturally relevant centers on understanding and addressing the unique cultural backgrounds, experiences, values, and perspectives of specific cultural groups or communities. The purpose is to cultivate meaningful connections and inclusivity within these cultural groups by acknowledging and affirming their identities and lived experiences. For example, tailored educational materials reflecting the language and traditions of a specific cultural group, healthcare practices sensitive to cultural beliefs, and media representation that celebrates diverse cultural identities. Culturally relevant refers to information, materials, practices, or approaches that are designed or chosen specifically to resonate with and be meaningful to a particular cultural group or community. Culturally relevant content acknowledges and respects the cultural backgrounds, experiences, values, and perspectives of the target audience. It aims to foster engagement, understanding, and connection by reflecting and affirming the cultural identities and lived experiences of individuals within that group. Culturally relevant content often incorporates culturally authentic elements, such as language, symbols, traditions, and narratives, to ensure that it is relatable, accessible, and impactful for the intended audience.

2.b. Question 2: What are promising strategies for handling sensitive and difficult conversations in a data science classroom?
To adequately and accurately introduce students to data science and the data sets upon which it is built, it may be necessary to bring up sensitive topics or new topics that may not be known to the students or the instructor. As students and instructors have different attitudes, values, and experiences, many may be new to these topics and may not know how to navigate the conversation with empathy. At times, this can lead to conversations that can upset some students, making them uncomfortable and, at times, angry and defeated. However, to understand the complexity of data science and the context of the data sets upon which data science is based, these conversations need to be conducted.
Advanced preparation by faculty members can help make these conversations more successful and reduce potential student stress and misunderstandings. Faculty members should consider “establishing community agreements- sets of principles/rules that everyone agrees to (i.e., rules of conduct, guidelines for discourse/language, etc..,” and have prepared strategies to deal with any students who become “disingenuous and disrespectful.”
Responsibly addressing historical and generational trauma that can surface in response to certain module topics is crucial when having sensitive and difficult conversations in data science. Research has shown that students from marginalized communities may carry the weight of past oppression and discrimination, which can profoundly shape their experiences and perspectives [17]. Faculty should be mindful that discussions around data and its applications may inadvertently trigger painful memories or feelings for these students. By acknowledging the potential for historical and generational trauma, faculty can create a more inclusive and compassionate learning environment [18]. Faculty should create a “safe space” policy and practice for class discussions and student engagement with sensitive topics.
By implementing certain strategies for handling sensitive and difficult conversations, faculty can create a classroom culture that prioritizes open, respectful, and productive dialogue on sensitive topics, enabling students to navigate the complexities of data science with greater depth and understanding. Establishing ground rules is essential for creating a respectful environment that can cultivate open dialogue on sensitive or new topics. Collaboratively developing norms like active listening, respecting opinions, and maintaining confidentiality can set the tone for productive discussions. Encouraging psychological safety is crucial, empowering
students to share their perspectives without fear of judgment or retaliation. The faculty instructor can model vulnerability and acknowledge biases, which can make students more forgiving and open to challenging dialogues.
Trust and respect are paramount for building a classroom climate that allows for sensitive and difficult conversations. Faculty should strive to create a culture that values inclusivity, diversity, and psychological safety [19]. This can be achieved by modeling vulnerability and being transparent about their own biases and limitations in understanding certain topics. Faculty can encourage “students to set their methods for discussion, building a sense of shared ownership and responsibility.” When contentious topics come about, faculty should be prepared to redirect the conversation back to the core learning outcomes while also validating the students’ perspectives and experiences. Students need to feel empowered to engage in sensitive topics and develop a deeper understanding of the complexities inherent in data and its interpretation. This is done by creating a culture of trust, which often starts before students walk into the classroom, with them understanding the history of the past.
Additionally, promoting a sense of shared ownership in the classroom can help build respect and engagement. Encouraging students to take an active role in leading discussions and interpreting data can empower them to take a more active stake in the learning process. As an ongoing process, faculty should regularly solicit feedback from students on the classroom environment and their comfort level with the discussions. This feedback can then be used to continually refine and adapt their approaches, ensuring that the learning space remains inclusive and responsive to the needs of all students.
Specific instructional strategies include:
• Set a goal to have a classroom where students and the instructor can have a respectful conversation where people can have different perspectives and understandings, not where everyone agrees but where people challenge their thinking and assumptions.
• If analyzing a sensitive dataset, use certain techniques that allow students to lead and contribute to the discussion while listening and respecting their interpretations. Provide gentle redirections as needed.
• If the discussion leads to a heated debate, put examples on the board of mathematics examples and equations as a way to remind everyone of the focus.
• Reinforce that the goal is to have a classroom where students and the instructor can have a “respectful conversation where people can have
who have expertise in data want to teach this type of course or get involved. As such, there should be flexibility in the expectations of who should teach the course.
After faculty members have been identified, it may be useful to gain an understanding of their expertise in data science and African diaspora studies. Given the transdisciplinary nature of the course, faculty could come from multiple disciplines and may need to upskill to ensure knowledge in the multiple facets of the course. While not all professors have complete knowledge before leading a course, key areas may aid in the preparation. If faculty has data science preparation, they might not have had the preparation to teach African diaspora studies and vice versa. Upskilling in these domains is key, and faculty members should ask themselves if they have the capacity and the bandwidth to meet the needs of teaching the course properly.
While the demographic characteristics of the faculty and students may not be resonant with those of the African diaspora, this course presents a novel opportunity to uncover nuances of cultural diversity through data. Having cultural competency and the ability to learn and demonstrate an understanding of different cultures will equip the faculty to lead the students through data projects that are informed by the various cultures that exist in the African diaspora. It is the role of the faculty member to ensure that the classroom is a safe space, and as such, cultural sensitivity must be utilized, e.g., in selecting the topics, leading classroom discussions, and assessing student work.
Being open to faculty collaborations to develop and possibly teach the course could enhance the student learning experience. Since faculty who may teach it may have greater expertise in a strand of the course (e.g., data science, African diaspora studies, and programming), leveraging those from outside the discipline as either guest speakers or to inform the development of the course could bring much value. In this case, the faculty member teaching the course must possess a degree of openness to learn, and if co-teaching is an option, a team-oriented approach should be taken. These collaborations could also recruit and prepare additional faculty members to teach this course so that it does not fall as a burden to one individual.
Once faculty have been identified, the immediate need for buy-in, institutional support, and funding becomes apparent. These supports, external resources, and faculty colleagues are crucial for advancing culturally relevant data science pedagogy. As an initial step, aligning the course with the institution’s mission, vision, and strategic plan may assist with institutional buy-in and support. Acknowledging and communicating the work already occurring within the institution or department may
also aid efforts. Teaching culturally relevant topics, such as a special topics course, may also be essential for garnering departmental buy-in.
Funding is another important resource needed to advance this work. Data science grants, based on collaborative efforts, can be used for various disciplines. Potential funding sources could be from both private or public foundations. However, the most valuable resources identified were faculty colleagues. The potential for multidisciplinary collaboration, both on the micro and macro levels, can lead to developing program curricula, best practices, and a public repository of programs/projects. A community of colleagues can prove instrumental in preparing class logistics, improving programs, and collaborating across university systems.
Multilevel institutional support is crucial to advancing the course. This support assists in identifying a core team for program development, creating sustainable equity and equality educational experiences, challenging the status quo, and creating an environment for new approaches and challenges the status quo. For example, a participant stated, “STEM faculty can help with the design of Python markdown scripts for specific modules” as an example of institutional support.
Professional development has been identified as an essential resource. Respondents expressed a need for digital resources, an instructor toolkit, and ongoing training workshops. Workshop topics included the data science life cycle and relevant data science pedagogical approaches. Finally, advisory boards and knowledge of relevant state statutes and terminology may prove helpful when navigating through difficult local/state political climates.
Content from resources, tools, and strategies for teaching about and advancing relevant data science in higher education. Freely available resources, such as Open Educational Resources (OER) or a university library, can provide key resources and datasets related to the African Diaspora. This content should be related to content analysis and communication, computer programming, and data sets. Finding appropriate data sets to conduct data investigations may include curated data sets that cover topics of interest to students.
The analysis and communication in the classroom contribute to promising instruction on data science instruction, including training students to understand the data science process, teaching them to “read, digest, and reflectively respond to data,” helping students to creatively tell the story of the data using visualization programming, and incorporating a “diverse range of scholars for classroom discussion and reading assignments.

Computer programming languages can vary in preference and can vary over time due to workforce and institutional demands. While Python and R are commonly used in data science, the learning of programming to meet the level of student preparation must be crafted with care. Creating engaging, in-class, interactive programming exercises can help students gain familiarity with coding, and the visualizations contribute to the understanding of the data. Depending on the student's prerequisite knowledge, there may be a review of data science and African diaspora topics to transition smoothly into the module. The incorporation of graphs, diagrams, and visuals into the modules, as well as the historical background of coding, can enhance the learning experience. Including foundational terms commonly used in programming (e.g., algorithm, data structures, formatting, pseudocode, etc.) helps build students' understanding.
Lectures should extend students' skills in mathematics and statistics, along with programming (e.g., R or Python). Providing hands-on, in-class activities for students to practice interpreting, writing, and debugging coding helps students gain confidence to explore the topics computationally. Incorporating team learning and project-based assignments motivates peer learning. Module assignments facilitate the development of student critical thinking skills. The following is a list of resources to consider:
1. Develop measurable course objectives, module objectives, and student learning outcomes that align with the learning activities and assessments specific to African diaspora
2. Incorporate metacognitive activities to introduce coding.
3. Provide context for the data set to assist with developing students’ critical thinking skills.
4. Scaffold activities and mini-lesson, just-in-time strategies to build student coding confidence.
2.c.ii. Instruction and Pedagogy. There are several key aspects of instruction and pedagogy in culturally relevant data science. For the assessment, using pre- and post-surveys to evaluate student progress and learning can inform the instructor of student progress and what the students are learning. To be culturally competent or sensitive, instructors should be mindful of each of their students and foster sensitive conversations around race, especially in diverse or homogeneous classrooms. Inviting multiple perspectives can help the class as a whole better understand the topic and also make sure that every student feels heard and valued. Leveraging faculty colleagues as a resource in navigating sensitive topics can equip the instructor to create a ‘safe space’ in the classroom. One method would be to start by setting ground rules for decorum in the discussion, posing relevant questions, and

then choosing the appropriate tools (e.g., SPSS, R, Python) to uncover data insights. Guiding students through culturally relevant content and hands-on learning experiences encourages critical thinking through the practical application of concepts.
2.c.iii. Strategies. Securing institutional buy-in is crucial for the sustainable implementation of a new curriculum. Fortunately, the integration of culturally relevant data science often aligns with the university’s mission. Explicitly providing this alignment can garner institutional buy-in and support to expand the curriculum and infrastructure. Gaining department-level backing from chairs, deans, and other colleagues is necessary to challenge the status quo and shift toward new approaches. External factors, such as state policies and political contexts, require careful navigation and could potentially supported by external advisory boards to connect learning to opportunities and workforce preparation.
Collaboration across disciplines and universities strengthens multidisciplinarity, which can be supported by team teaching. Although team teaching may be a challenge to implement, it provides an opportunity to cross-pollinate disciplinary areas to enhance student learning. To support the teaching of the course and facilitate collaborations, faculty development through workshops can make an impact on pedagogy and strengthen collaborations between faculty. Securing funding through grants and offering digital resources can help launch and support these efforts.
4. APPENDIX
4a. Datasets for Teaching and Learning Data Science
Data Topic Dataset Name and Description
African DiasporaEnslavement Counts
Documentation
African DiasporaEnslavement Counts
Cause of death in Black Americans over time. Data on causes of death among Black Americans.
Documentation Slave Voyages. Database detailing historical slave voyages.
African DiasporaPopulation Pew Research Center, Black Americans. Research data on Black Americans from Pew.
African DiasporaPopulation
Child Welfare Data
African Population and Migration Dataverse. Dataset on African population and migration trends.
Child Opportunity Index database (grapple), Data for a diverse and equitable future. Provides data on child opportunity across various dimensions to assess equity in opportunities for children.
Child Welfare Data Kids Count Data Center. Database providing child well-being data by state. Data on children in foster care broken down by race/ethnicity.
Child Welfare Data National Center for Juvenile Justice. Data dashboard on juvenile justice and disparities.
Crime Data
FBI's Crime Data Explorer (CDE). FBI's portal for crime statistics.
Education AP Score distribution. Data on AP exam scores and distributions.
Environmental Data Air quality changes (1998-2019). Maps based on changes in air quality particulates. (1998 - 2019)
Health Data
Covid-19 Data Tracker. Data on COVID-19 impacts by income and race in the USA.
Police Shooting Data Datasets from Kaggle related to police shootings in the USA. Multiple datasets on
Related to Teaching Undergraduate Students Introduction to Data Dataset Link
Provides historical health data, relevant for teaching epidemiology, health disparities, and historical analysis.
Important for teaching historical analysis, transatlantic slavery, and global economic impacts.
Provides demographic insights, useful for teaching social science research methods, demographics, and public opinion analysis.
Useful for teaching demographic analysis, migration patterns, and global development issues.
Useful for understanding social equity issues and can be used to teach data visualization and equity analysis.
Offers state-level child well-being metrics, suitable for teaching community health, policy analysis, and data visualization. Relevant for teaching social work analytics, child welfare policy, and racial disparities in foster care systems.
Useful for teaching criminal justice analytics, juvenile justice policy, and data-driven decisionmaking.
Offers detailed crime data useful for teaching criminal justice analytics, law enforcement policy, and data visualization.
Useful for teaching statistical analysis and educational equity, comparing performance across demographic groups.
Relevant for teaching environmental data analysis, pollution impacts, and spatial correlation studies.
Provides insights into the pandemic's effects on different demographics, useful for teaching public health data analysis and socioeconomic
Provides various perspectives on police violence, useful for teaching data cleaning, analysis, and
https://enslaved.org/data/
https://www.slavevoyages.org /
https://www.pewresearch.org /tools-and-resources/
https://dataverse.harvard.edu /dataverse/WH_AfricanPopM igration
https://data.diversitydatakids. org/dataset/coi20-childopportunity-index-2-0database
https://datacenter.aecf.org/da ta#USA/1/0/char/1
https://ncjj.org/AFCARS/Di sproportionality_Dashboard.as p
https://cde.ucr.cjis.gov/LAT EST/webapp/#/pages/home
https://apstudents.collegeboar d.org/about-ap-scores/scoredistributions
https://sedac.ciesin.columbia. edu/data/set/sdei-globalannual-gwr-pm2-5-modismisr-seawifs-aod-v4-gl-03
https://covid.cdc.gov/coviddata-tracker/#datatrackerhome
https://www.kaggle.com/data sets/ahsen1330/us-police-
police shootings from Kaggle. ethics in research.
Policing Data Stanford Open Policing Project. Project analyzing police practices and racial disparities.
Policing Data
Policing DataComplaints
Policing DataPolice Shooting
State of California DOJ Racial and Identity Profiling Act (RIPA) Stop Data collection on police stops
NYPD Misconduct Complaint Database. Database of misconduct complaints against NYPD officers.
US Police Shootings (Kaggle). Dataset compiling information on police shootings in the USA.
Policing DataPolice Shooting Police shootings database. Database compiled by the Washington Post on police shootings.
Policing DataPractices Racial and Identity Profiling Advisory Board (RIPA) reports on racial profiling and police practices.
Policing DataTraffic Stops Data on traffic stops in Connecticut.
Population Data
Population Data
Substance Use/Abuse Data
Policing DataTraffic Stop
American Community Survey, Median Income (Inflation-adjusted). Census data on median income adjusted for inflation.
US Census Bureau. Various tables from the US Census Bureau.
The opioid crisis - geographic and numerical values focused on urban or rural regions. Data analyzing the opioid crisis by geographic area.
North Carolina Traffic Stops, Department of Political Science. Traffic stop data for North Carolina was recorded by the Department of Political Science at UNC-Chapel Hill.
Provides datasets on traffic stops and policing practices, useful for teaching data ethics, bias detection, and social justice issues. Requires R or Python to download.
Relevant for teaching state-level policy analysis, civil rights, and data-driven policy evaluation.
Offers insights into police misconduct, ideal for teaching data analysis, ethics, and public policy implications.
Relevant for studying race and policing issues, suitable for teaching statistical analysis and social justice topics.
Provides detailed data on police shootings, suitable for teaching data journalism, ethics, and social justice issues.
Offers insights into policing practices, suitable for teaching criminal justice policy and civil rights issues.
Provides local-level traffic stop data, useful for teaching data analysis, law enforcement policy, and civil rights issues.
Provides socioeconomic indicators, useful for teaching economic analysis, inequality, and policy impacts.
Provides comprehensive demographic and socioeconomic data for teaching data analysis and policy research.
Relevant for teaching spatial analysis, public health data, and socioeconomic impacts of health crises.
Provides real-world data on traffic stops, which can be used to teach data analysis, bias detection, and social justice issues.
shootings,%20https:/www.kag gle.com/datasets/mrmorj/dat a-police-shootings
https://openpolicing.stanford. edu/data/
https://openjustice.doj.ca.gov /exploration/stop-data
NYPD Misconduct Complaint Database Updated
https://www.kaggle.com/data sets/ahsen1330/us-policeshootings
https://washingtonpost.com/ graphics/investigations/police -shootings-database/
https://oag.ca.gov/ab953/boa rd/reports
http://trafficstops.ctdata.org/
https://data.census.gov/cedsci /table/ACSST1Y2022.S1903? q=Income%20and%20race&g =010XX00US$0400000&y=2 022&moe=true&tp=true
https://data.census.gov/cedsci /table?q=United%20States
https://www.opencasestudies. org/ocs-bp-opioid-ruralurban/#Motivation
https://fbaum.unc.edu/traffic. htm
4.b. Agenda and Structure of the Workshop on Enhancing Data Science
Education by Leveraging Data Sets from the African Diaspora
Workshop Website: https://www.accelevents.com/e/datadiaspora
Agenda (Times in EDT)
11:30 –12:15 pm
12:151:00 pm
1:00 –1:45 pm
1:45 –2:30 pm
2:30 –3:15 pm
3:15 –3:20 pm
3:20 –3:35 pm
Group Sharing Report-backs and discussion in a large group
Panel 2: Benn, Khadjavi, A. N. Washington
Long Break
On culturally relevant data in relation to bias, ethics, and contemporary issues that impact Black America
Panel 3: Wall Rice, Winston Cultural sensitivity in data science pedagogy
Breakout Sessions
Short Break
Closing & Feedback
What are promising strategies for handling sensitive and difficult conversations in a data science classroom?
3:35 –4:00 pm
Preview Day 3 and explain “homework” (write 1-2 page reflection paper on reactions to workshop dialogue on handling sensitive racial topics in the data science classroom). Complete a 3-question feedback survey at http://www.campbellkibler.com/AU/AU-ExploreTeach.html A summary of the results will be shared tomorrow morning.
Networking / Social Session Optional
Wednesday, May 18, 2022: Exploring Promising Tools and Strategies to Enhance Culturally Relevant Pedagogy for Data Science
11:00 –11:30 am
11:30 –12:15 pm
Breakout Sessions
Review Day 2 homework on sensitive conversations
Group Sharing Report-backs and discussion in a large group
12:15 –1:00 pm Long Break
1:00 –1:45 pm
1:45 –2:45 pm
Panel 4: Black, Bressoud, and Williams
Tools and strategies for teaching culturally relevant data science (e.g., pedagogical approaches, software tools, platforms, institutional barriers to curricular innovations, and faculty development/training)
Breakout Sessions What is needed to teach culturally relevant data science?
Time Title Description
2:45 –
3:00 pm
3:00 –
3:45 pm
3:45 –
4:00 pm
Short Break
Group Sharing
Closing
Report-backs and discussion in a large group
Discussion of potential next steps and completion of the workshop evaluation at http://www.campbellkibler.com/AU/AU-Evaluation.html
Organization: The three-day discovery workshop Enhancing Data Science Education by Leveraging Data Sets from the African Diaspora will virtually convene a cohort of 40 participants tasked with interacting to explore and discover key topics and questions that will inform future research and collaboration around how to better practice culturally relevant data science education at the undergraduate level.
Breakout Sessions and Collaboration: Much of the collaboration in the workshop will occur in breakout sessions with around eight participants each. Attendees will receive pre-curated "Google document worksheets" to help them generate, refine, and assess ideas and options as they explore workshop topics. This tool will allow attendees who have access to a computer to input their ideas directly into a collaborative document. Each breakout session will be staffed by co-facilitators and a graduate student scribe. Each breakout group will be curated to reflect the diverse representation.
Keynote: The keynote aims to motivate the participants to rethink data science in the context of culturally relevant data science and the experience of those of the African diaspora. In a motivational and conversational presentation style, the keynote will aim to inspire the participants to explore the interplay of data science and the experience of those in the African diaspora and convey why this is important. The keynote will present for half of the time and then enter into a 'fireside chat' with a designated moderator to do a deeper dive with the participants.
Panels: The panels aim to inform the participants about specific best practices, resources, literature, and other items to consider. Each panel will have 2-3 presenters where each presenter will talk for 10 minutes, and then with a designated moderator, engage with the audience in a discussion style to explore the topics further with the participants via a Q&A.
Agenda: Workshop sessions will focus on the following themes:
● Teaching data validity and fairness versus bias, including incorporating reliable data sets that are relevant to African diasporic identity
● Brainstorming promising practices and exploring the benefits of incorporating culturally relevant data science in undergraduate education
● Identifying and probing data science topics that are highly relevant to Black America
● Exploring strategies for how to have difficult, "brave" conversations (e.g., discussing race and racism) in the data science classroom
● Identifying needs and opportunities for teaching culturally relevant data science to undergraduates (e.g., pedagogical approaches, software tools, platforms, faculty development/training, institutional barriers to / resources for curricular innovation)
Speakers:
● Dr. Nathan Alexander, Assistant Professor of Data Science and Interdisciplinary Studies, at Morehouse College, teaches courses in mathematics, computational methods, and education. His research explores the development of critical and justice-oriented practices in quantitative literacy development, with a particular focus on Black history and Afrofuturism. At Morehouse, Dr. Alexander directs the Quantitative Histories Workshop, a communitycentered teaching and learning lab for students, faculty, and community members in the Atlanta University Center.
● Dr. Emma K. T. Benn, is an Associate Professor in the Center for Biostatistics and Department of Population Health Science and Policy and Associate Dean of Faculty Wellbeing and Development at the Icahn School of Medicine at Mount Sinai. She is also the Founding Director of the Center for Scientific Diversity which applies a multidisciplinary research-driven approach to develop, implement, and evaluate best practices for promoting and sustaining inclusive excellence and increasing the recruitment, retention, research success, and advancement of underrepresented faculty and trainees in the biomedical research workforce. She teaches a graduate-level course, Race and Causal Inference, aimed at identifying effective causal targets for eliminating health disparities.
● Dr. Jason Black, Associate Professor of Information Systems at Florida Agricultural and Mechanical University, researches business analytics / big data, mobile computing, and educational technology. As the co-lead of the HBCU Data Science Consortium, he promotes collaboration and support of data science across the HBCU community.
● Dr. David Bressoud, Former Director, of the Conference Board of the Mathematical Science (CMBS), an umbrella organization consisting of nineteen professional societies to promote research, improve education, and expand the uses of mathematics. As former President of the Mathematical Association of America, he has led initiatives on the role of calculus and most recently, is delving into the role of data science in undergraduate education.
● Dr. Nicholas Horton, Beitzel Professor in Technology and Society (Statistics and Data Science) at Amherst College, researches bridging the gap between theory and practices in interdisciplinary statistics and data science. His work includes expanding data science education, and he has been involved in initiatives through the National Academies.
● Dr. Monica Jackson, Interim Deputy Provost, Dean of Faculty and Professor of Statistics, at American University, researches spatial statistics and disease surveillance. She co-directs
the Summer Program in Research and Learning (SPIRAL) to train undergraduate students to conduct research and statistics with a focus on training underrepresented minorities.
● Dr. Lily Khadjavi, Professor of Mathematics, at Loyola Marymount University, research interests lie in the intersections of mathematics and social justice, including policing and the issue of racial profiling. She is co-editor of the book "Mathematics for Social Justice: Resources for the College Classroom".
● Dr. Victoria Robinson, Lecturer at UC Berkeley in Ethnic Studies and Women’s Studies, teaching courses addressing race and ethnicity in the United States and global female migrations. Her most recent area of research addresses the gendering of post-industrial return migrations to the Caribbean.
● Dr. A. Nicki Washington, Professor of the Practice in Computer Science and Gender, Sexuality, and Feminist Studies (Duke University), focuses on identity in computing, specifically in understanding how identity impacts and is impacted by computing, as well as measuring and improving the cultural competence of faculty, staff, and students to better engage and retain marginalized students.
● Dr. Cynthia Winston-Proctor, Professor of Psychology at Howard University, as well as the P.I. of the Identity and Success Research Laboratory. Her interdisciplinary narrative personality psychology research explores applications to culturally relevant STEM educational design, professional development of women in the workplace, emotional intelligence, behavioral cybersecurity, psychology engineering, A.I. narrative design, African ancestry tracing, and the psychological well-being of Black women. She also is a P.I. of the NSF Broadening Participation Research Center for the Development of Identify and Motivation of African American Students in STEM" funded by NSF 2010860.
● Dr. Talithia Williams, Associate Professor of Mathematics at Harvey Mudd College, develops statistical models that emphasize the structure of data, and she makes statistics and data science understandable to a wide audience. As the former Associate Dean for Research and Experiential Learning, she provided opportunities for faculty to develop research and led efforts to broaden the summer research and experiential learning program.
Breakout Group Participants, Unit/Department, Institution (* denotes HBCU; + denotes speaker)
Facilitator: LaTanya Brown-Robertson, Bowie State University*
Assistant Facilitator: Bolametiren Akinlaja, Bowie State University*
● Barbara Harris Combs, Sociology and Criminal Justice, Clark Atlanta University*
● Ranthony Edmonds, Mathematics, The Ohio State University
● Lethia Jackson, Technology & Security, Bowie State University*
● Celeste Lee, Sociology, Spelman College*
● Chuang Peng, Mathematics, Morehouse College*
● Lei Qian, Computer Science, Fisk University*
Facilitator: Moses Garuba, Howard University*
Assistant Facilitator: Victoria Grase, Clark Atlanta University*
● Leslie Collins, Behavioral Sciences, Fisk University*
● Marionette Holmes, Economics, Spelman College*
● Nicholas Horton+, Mathematics and Statistics, Amherst College
● Binod Manandhar, Mathematical Sciences, Clark Atlanta University*
● Mandoye Ndoye, Electrical and Computer Engineering, Tuskegee University*
Facilitator: Sajid Hussain, Fisk University*
Assistant Facilitator: Magana Kabugi, Fisk University*
1. Jason Black+, School of Business / Information Systems and Operations Management, Florida A&M University*
2. Elycia Daniel, Sociology/Criminal Justice, Clark Atlanta University*
3. Allen Hillery, Academic, Macaulay Honors College*
4. Yvonne Phillips, Computer Science/Kinesiology, Morehouse College*
5. Carmen Wright, Mathematics and Statistical Sciences, Jackson State University*
6. Bo Yang, Computer Science, Bowie State University*
Facilitator: Eric Van Dusen, University of California at Berkeley
Assistant Facilitator: Kelechi Nnebedum, University of California at Berkeley
● Nathan Alexander+, Experiential Learning and Interdisciplinary Studies, Morehouse College*
● Michelle Juarez, DNA Learning Center, Cold Spring Harbor Laboratory
● Ivis King, Social Work, Clark Atlanta University*
● Widodo Samyono, Mathematics, Jarvis Christian College*
● Alfred Watkins, Computer Science, Morehouse College*
● Azene Zenebe, Information Systems, Bowie State University*
Facilitator: Eboni Dotson, Morehouse School of Medicine*
Assistant Facilitator: TaMiko Condoll, Morehouse School of Medicine*
● Viveka Brown, Mathematics, Spelman College*
● Michelle Homp, Mathematics, and Center for Science, Mathematics & Computer Education, University of Nebraska - Lincoln
● Magana Kabugi, English, Fisk University*
● Antwain Leach, Political Science/African American Studies/General Education, Fisk University*
● Felesia Stukes, Computer Science & Engineering, Johnson C. Smith University*
● Sairam Tangirala, School of Science and Technology, Georgia Gwinnett College
Facilitator: Jerry Volcy, Spelman College*
Assistant Facilitator: Kayla Partee, Clark Atlanta University*
● Debzani Deb, Computer Science / Center for Applied Data Science, Winston-Salem State University*
● John Harkless, Chemistry, Howard University*
● Lily Khadjavi+, Mathematics, Loyola Marymount University
● Qingxia Li, Mathematics and Computer Science, Fisk University*
● Shannell Thomas, Behavioral Sciences and Human Services, Bowie State University*
Brief Overview of Workshop Participants
Organizing Team:
● Co-Chair, Talitha Washington, Director of the Atlanta University Center Data Science Initiative; Professor of Mathematical Sciences at Clark Atlanta University
● Co-Chair, Jerry Vocy, Brown Simmons Professor of Science and Director of the Innovation Lab at Spelman College
● LaTanya Brown-Robertson, Professor of Economics at Bowie State University
● Moses Garuba, Associate Dean for Academic Affairs, Associate Director of the Data Science and Cyber Security Center, and Professor of Computer Science at Howard University
● Sajid Hussain, Associate Vice Provost for Innovation & Information Technology and the Discipline Coordinator of Data Science at Fisk University
● Torina Lewis, Associate Executive Director for Meetings and Professional Services at the American Mathematical Society
● Eric Van Dusen, Interim Director and Lecturer at the Data Science Program at the University of California, Berkeley
4c. Data and the African Diaspora Writing Retreat
Held on June 4 – 5, 2024, 9:00 am – 4:30 pm in Atlanta, Georgia, to synthesize feedback from the workshop. Participants used a shared Google Folder to do collaborative writing and analysis.
Agenda
Day 1: Report Preparation and Initial Drafting Time Facilitator
8:30 AM - 9:00 AM Sign-in, Technology set-up + Coffee and Cake Dotson
9:00 AM -9:40 AM Welcome and Overview 40 minutes Welcome & Introductions
DAD Writing Retreat Materials
Purpose
● How it started and where we’re going Overview of Workshop Tasks:
● Review qualitative analysis, consensus report, and feedback
● Write a report on the findings
9:40 AM - 10:15 AM Outlining the Final Report
● Structure and components
● Report Outline
● The approach of qualitative analysis and data sources
● Assigning writing teams
o Question 2 writeup
King and Phillips Washington
● Introduction of content from 2022 DAD Workshop breakout groups’ 20 minutes King
10:15 AM - 10:30 AM Break Dotson
10:30 AM - 12:00 PM Writing Teams: Qualitative Analysis - Data Review
● Data Review and Organization by Teams
● Writing Session #1 Tasks Keywords/Code Framework
Phillips
12:00 PM - 1:00 PM Lunch (Working) Dotson
● Teams provide updates and feedback 60 minutes Washington
1:00 PM - 2:30 PM Writing Teams: Qualitative Analysis
● Writing Session #2 Task 90 minutes Phillips
2:30 PM - 2:45 PM Break Dotson
2:45 PM - 4:00 PM Writing Teams: Findings and Results Compilation
● Discuss preliminary findings from qualitative analysis
● Writing Session #3 Task Collaborative writing of the findings and results sections 5 minutes 60 minutes King and Phillips
4:00 PM - 4:30 PM Day 1 Wrap-up and Next Steps
● Writing Teams share out progress & to-do’s
● Optional: Gather/Upload additional articles and other resources
o Relevant literature on data science education and African Diaspora data sets
King and Washington
Day 2: Draft Refinement and Finalization
8:30 AM - 9:00 AM Check-in, and Coffee and Cake
9:30 AM - 10:00 AM Recap and Review
● Reflect on Day 1
● Writing Teams share reflections on Day 1
● Discuss additional articles and other resources
● Review of tasks to complete
Time Facilitator
minutes Washington
10:00 PM - 12:00 PM Writing Teams: Supporting Visuals in the Report
● Develop ideas for figures, graphs, charts, etc.
o E.g., workshop questions, speakers, homework, participant feedback
o Writing Session #4 Task
minutes King
12:00 PM - 1:00 PM Lunch (Working) Dotson
● Interpret the results: Implications and recommendations
● Share ideas on the discussion section and implications, e.g., future courses and research 60 minutes Phillips
1:00 PM - 1:30 PM Lessons Learned + Discussion
DAD Course: Lessons Learned in the AUC
● Clark Atlanta University
● Morehouse College
● Morehouse School of Medicine
● Spelman College
● DAD Course Information 30 minutes King and Phillips
1:30 PM - 1:45 PM Break
1:45 PM - 3:30 PM Writing Teams: Finalizing the Report Sections
● Integrate all sections into a cohesive draft
● Assign final editing tasks
● Writing Session #5
3:30 PM - 4:00 PM Dissemination Plan Development
● Develop a dissemination plan for the report
o Who? What? How? Where?
● Identify target audiences and dissemination channels
● Plan the virtual webinar for dissemination
● Final edits and review of the consensus report
minutes
minutes King
● Writing Session #6 30 minutes Washington and Phillips
4:00 PM - 4:30 PM Next Steps
● Outline of next steps (e.g., report submission, engaging past participants, and finalization of the report)
● Closing remarks and acknowledgments 30 minutes Washington and King
Attendees
First Name
Ivis Renee King, Lead Facilitator
Yvonne Phillips, Co-Lead Facilitator
Adrienne Avery
Patricia Campbell
Eboni Dotson
Angelita Howard
Torina Lewis
Binod Manandhar
Nyya Parson-Hudson
Kenzy Scott
Jamal Ware
Talitha Washington
Jackie Williams
Clark Atlanta University
Morehouse College
Atlanta University Center Consortium
Campbell-Kibler Associates
Atlanta University Center Consortium
Meharry Medical College
Prairie View A&M University
Clark Atlanta University
Clark Atlanta University
Atlanta University Center Consortium
Atlanta University Center Consortium
Clark Atlanta University
Clark Atlanta University