Data Science Transfer Pathways

from Associate’s to Bachelor’s Programs

Benjamin S. Baumer (with Nicholas J. Horton)

Data and AI Meeting of RI Regional Academic Institutions
September 16, 2022

DSC-WAV

• Wrangle, Analyze, Visualize

The Data Science Corps WAV

• Experiential Learning with Local Community Organizations

• 4 years, 9 institutions, $1.2 million budget

• 25-30 students per semester doing projects, 5 cohorts begun January 2020

• working in parallel with three community colleges

• Thanks to NSF #1924017 and #1923388 for funding this work

DSC-WAV goals

1. Workforce development for undergraduates
   • focus on diversifying the workforce
   • 74% student participants female or non-binary
   • ~46% student participants non-white
2. Building the pipeline
   • Two-year college students to bachelor’s
   • MTH 190 at HCC

Goal 1: Partners

Goal 1: Papers

• Horton et al. (2021), The Data Science Corps Wrangle-Analyze-Visualize program: building data acumen for undergraduate students
  Harvard Data Science Review
• Legacy et al. (2021), Facilitating Team-Based Data Science: Lessons Learned from the DSC-WAV Project
  Special issue on “Data Science Education Research”
  Foundations of Data Science

Goal 2

• building data science programs at community colleges

• constructing transfer pathways

Road map

1. Motivate the importance of data science
2. Familiarize institutional leadership with options
3. Prepare faculty to teach data science
4. Identify barriers to pursuing data science programs
5. Share resources and best practices
6. Explore next steps

Community Colleges

“the keystone for the nation’s plan to help more people earn a postsecondary credential.”

–Blumenstyk (2021)

• substantial fraction of undergraduate students
• only affordable game in town
• critical to develop an educated workforce
• more representative of the population
• need for flexible and innovative articulation (transfer)

NASEM (2018)

• definition of data acumen
• importance of community colleges
• cooperation with four-year institutions

2YCDS Summit (2019)

• Examples of successful CC data science programs
• What is different at community colleges?
• How can we support faculty development, curricular initiatives

Transfer pathways

Goal: A2B pathways on website

• Mapped A2B pathways for data science

• BHCC -> UMD?
  mock-up!!

• HCC -> UMA?

Requirements

• Need a student-centered approach to foster success
• Need to support associate’s to workforce
• Need to support associate’s to transfer (today’s focus)
• What are the points of friction we need to smooth or eliminate?

Five points of curricular friction

1. A first course in data science
2. A second course in data science ()
3. A course in scientific computing, data science workflow, and/or reproducible computing
4. Lab sciences
5. Communication, ethics, and application domain

1: A first course in data science

In order for data science transfer pathways to work, community colleges must offer a first course in data science.

“To prepare their graduates for this new data-driven era, academic institutions should encourage the development of a basic understanding of data science in all undergraduates.”

–NASEM (2018)

1: first course (status)

• BHCC CIT-137 Intro BigData w/R & RStudio
  Michael Harris

• HCC MTH 190
  Nic Schwab

• subject of FacDev22 Workshop

• safe prediction: more on the way…

1: first course (transfer targets)

• UMass-Dartmouth DSC 101
  Donghui Yan
  UMass-Dartmouth Curriculum Flowchart

• Westfield CAIS 0105
  ??

• UMass-Amherst INFO 248

• safe prediction: more on the way…

2: Second course in data science

Cultivating a rich facility in data science requires repeated exposure: a single course is not sufficient for students to develop mastery.

• BA programs will include a 2nd course in data science
  • often taken during the sophomore year
• reinforce and extend fundamental skills
  • data wrangling, data visualization, statistical modeling, and predictive analytics.
• UMass-Dartmouth DSC 201
  David Koop

3: A course in scientific computing, data science workflow, and reproducible computing

A generic bachelor’s program in data science will include explicit instruction in how to advance science by computing with data in a reproducible, collaborative workflow.

• idiosyncratic

4: Lab science

Many of the existing STEM transfer options require two semesters of lab sciences as a component of their general education requirements.

• less germane for data science students?
• future data science infused lab course?

5: Communication, ethics, and application domains

Bachelor’s programs in data science include training in communication and ethics (what responsibilities to data scientists have to their users, customers, and society as a whole?, see Baumer et al, 2022).

• find ways to use or modify existing courses

Our paper

• Baumer and Horton (under review), Fostering and Simplifying Data Science Transfer Pathways in Massachusetts, Harvard Data Science Review

Thank you!