Lecture 1-1: Course Overview
DATA 510: Data Science Capstone
Orienting lecture for DATA 510: catalog goals, grading components and weights, milestone and weekly deliverable schedule, Data-Driven Scrum methodology, meta-project clusters, late policy, attendance and integrity expectations, and workload norms aligned with the published syllabus.
Learning objectives
Today’s objectives
What you will leave with
By the end of this session, you will be able to:
- Explain how the capstone grade is built from milestones, process habits, and communication deliverables (including where to find authoritative due dates).
- Describe team options, meta-project clusters, and why weekly DDS rituals exist.
- Locate major milestone due weeks on the semester calendar and what each milestone is meant to prove.
- Apply the late-work grace rule correctly (including known exemptions).
This deck summarizes the syllabus and methodology pages; Canvas remains authoritative when specifics differ.
Part 1: What this course is
DATA 510 in one paragraph
Over the semester you propose, plan, and execute a real data science project that integrates core MS program skills and produces a portfolio-quality artifact aligned with your career goals.
Projects should be consequential: plausible impact on an organization or problem domain. Grades reward steady progress, communication quality, and how you present and defend work (including feedback from industry judges where applicable).
Meeting logistics
Classes meet Monday, 6:00 PM to 10:00 PM, Ford 102.
Each evening typically splits into two blocks (6:00 to 7:30 PM and 8:00 to 10:00 PM). Expect a blend of seminar discussion, short lectures, group work, mentoring, and guests (the mix may shift with guest schedules).
Catalog framing
You will:
- Select data sources that are publicly available or publishable, with instructor approval early in the term (target data decisions inside the first two weeks alongside chartering and proposal work).
- Experience checkpoints that mimic consistent workplace deadlines, with structured peer and instructor feedback.
Prerequisites and materials
Prerequisites: DATA 501, DATA 502, DATA 503, DATA 504.
Minimum grade: A grade of C- or better is required for this course to count toward university credit.
Textbook: None. Readings and links will appear in Canvas.
Course objectives (official)
- Implement modern development practices that support publication and presentation of a data science project.
- Discuss recent developments in data science and their organizational and societal implications intelligently.
- Propose and defend a data-science-based solution to a problem of consequence.
- Communicate implications clearly to non-technical audiences, in person and in writing.
- Demonstrate readiness for professional interview settings on core data science topics.
Student learning objectives (high level)
You will practice how to:
- Navigate hiring and interview effectively (professional development sequence).
- Market your strengths through GitHub presence, the final write-up, and public artifacts.
- Plan a large research project end-to-end with realistic timelines (proposal and milestones).
- Execute on that plan across graded checkpoints (milestone submissions).
- Present findings interactively to technical and non-technical audiences (final poster session and rehearsals).
- Write for modern publication on the web (rough and final drafts).
- Collaborate so workloads split fairly when you choose team execution.
Student learning objectives map directly onto graded components such as professional development, milestones, and the final presentation and write-up.
Part 2: How your grade is built
Grade components at a glance
Canvas combines weighted components below. If anything here disagrees with a dated Canvas announcement or assignment page, follow Canvas.
| Component | Weight |
|---|---|
| Professional development activities | 10% |
| Weekly peer and meta-project evaluations | 5% |
| Project process and practices (includes Git workflow) | 5% |
| Technical writing practice | 5% |
| Project proposal | 9% |
| Data summary | 9% |
| Poster rough draft | 8% |
| Written report rough draft | 9% |
| Final poster presentation | 18% |
| Final write-up | 14% |
| Project website and dissemination | 8% |
| Total | 100% |
Letter grade cutoffs
| Average | Letter |
|---|---|
| 92.00+ | A |
| 90.00 to 91.99 | A- |
| 88.00 to 89.99 | B+ |
| 82.00 to 87.99 | B |
| 80.00 to 81.99 | B- |
| 78.00 to 79.99 | C+ |
| Average | Letter |
|---|---|
| 72.00 to 77.99 | C |
| 70.00 to 71.99 | C- |
| 68.00 to 69.99 | D+ |
| 62.00 to 67.99 | D |
| 60.00 to 61.99 | D- |
| 59.99 and below | F |
Ongoing coursework (what those percentages mean)
Professional development (10%). Every other week (or so) opens with structured career preparation (even if you are not job searching immediately).
Weekly peer and meta-project evaluations (5%). Short structured reflections on your team and the two peer projects in your assigned cluster (quality of feedback you gave, what you learned, risks you see).
Project process and practices (5%). Includes Git workflow, chartering, weekly summaries, and related habits. Git commit quality is reviewed weekly; grading for improvement incorporates feedback especially in the last five weeks of the term.
Technical writing practice (5%). In-class exercises with discussion. Partial credit only for makeup work if you miss the class session.
Project milestones (weighted artifacts)
Project proposal (9%). Short plan covering the research question, data sources, planned analysis, ethics notes, and feasibility. It is the default roadmap when scope needs to tighten.
Data summary (9%). Evidence that needed data are gathered and organized or pipelines run without ongoing manual babysitting. After this milestone, no further active ingestion work should be required.
Poster rough draft (8%). First end-to-end story on the wall; placeholders allowed, but the narrative arc should be complete enough to critique.
Written report rough draft (9%). First full draft of publishable prose across required sections. After this milestone, no further methodology or findings changes (editing and polish continue).
Final deliverables
Final poster presentation (18%). Conference-style poster session for peers, faculty, and industry guests; joint across cohorts on the last class evening.
Final write-up (14%). Polished, proofread permanent record; published and linked from portfolios.
Project website and dissemination (8%). Public-facing project materials (for example GitHub Pages) with clear links suitable for employers.
Rubric details live with each Canvas assignment; ask early if a component’s expectations are unclear.
Part 3: Methodology and teamwork
Team size and meta-projects
You may work solo or in a self-selected team of two or three.
Regardless of team size, every project sits in a meta-project cluster: your capstone is grouped with two other projects for the whole semester. You follow those peers closely and give structured weekly feedback (and receive it).
Higher bar for multi-person teams
Teams of two or three should carry noticeably higher scope, risk, or surface area than a comparable solo project (integration depth, evaluation rigor, division of labor, documentation).
Instructor approval of team scope happens at the project proposal. Proposals that look like solo scope with extra names should expect revision requests.
Data-Driven Scrum (DDS): why it exists
DDS is the course operating model for transparent, prioritized iteration tailored to data science.
It emphasizes:
- Collaboration and communication through visible boards and weekly narratives.
- Empirical iteration (short cycles of ideate, build, observe, analyze).
- Highest priority next through backlog refinement.
DDS flow (weekly rhythm)
- Brainstorm concrete backlog items (stories, spikes, experiments).
- Prioritize against data and modeling constraints.
- Create and refine artifacts tied to the top item.
- Observe results and reprioritize honestly.
Core artifacts include items, backlog, item breakdown board (IBB), task board, and the weekly progress report (often README updates plus snapshots when required).
Written weekly cycle (GitHub)
README or the filename specified in Canvas should routinely cover:
- Iteration review: what shipped and board snapshot.
- Retrospective: one process improvement for next cycle.
- Planning: which backlog items you pulled next and why.
In-class rituals
Standups and backlog refinement are in-class work, expected finished by the end of the evening session, including cross-team touchpoints inside your meta-project cluster.
Written follow-ups post by Canvas deadlines.
Meta-project reviews (what good feedback looks like)
Each week you:
- Read peer boards and weekly summaries.
- Give specific, kind feedback tied to artifacts (not personalities).
- Incorporate useful signals into your backlog; clarify politely when peers misread context.
Strong comments cite visible evidence, separate curiosity from blocking risks, and surface ethics and engineering concerns early (PII, leakage, reproducibility).
Part 4: Policies and expectations
Late work
Each deliverable has a 24-hour grace window after the stated deadline with no penalty.
Beyond that you lose 20% of that deliverable’s points per day late.
Exceptions (no grace window): Final presentations (single shared slot), writing assignments tied to technical writing, and Git-related assignments as labeled in Canvas.
Attendance and participation
Participation matters. You are responsible for making up missed material.
More than three unexcused absences may result in a failing grade for the course.
Incomplete grades
Incompletes are only considered for prolonged illness or family emergencies that remove you from campus for an extended period late in the term.
They are not granted for falling behind due to motivation, misunderstanding, or time management alone. If you are worried, come talk early.
Regrading
Email a written petition before the last class period. Explain why a regrade is warranted with references to course standards. Decisions after review are final.
Integrity and classroom conduct
Cheating and plagiarism undermine capstone credibility; penalties range from assignment failure to course failure.
Constructive behaviors build trust; disruptive behaviors (including unauthorized recording) interfere with learning and may be addressed under university policies.
Time on task and access needs
Expect roughly 8 to 12 hours of out-of-class work most weeks for a four-hour weekly meeting (more when deadlines cluster).
Contact Accessible Education Services if you need accommodations; talk to the instructor early if course design creates barriers.
Part 5: Semester schedule snapshot
Weeks 1 to 7 (topics abbreviated)
| Week | Date | Block 1 (6:00 to 7:30 PM) | Block 2 (8:00 to 10:00 PM) | Deliverables due end of class unless noted |
|---|---|---|---|---|
| 1 | 5/11 | Syllabus workshop; Research vs. EDA | Project brainstorming | Project brainstorming survey |
| 2 | 5/18 | Technical writing I | Chartering explained; chartering working session | Technical writing output; project charter |
| 3 | 5/25 | Git for project work | Project working session | Weekly summary report |
| 4 | 6/1 | PD: alum panel | Project working session | Weekly summary; project proposal |
| 5 | 6/8 | Technical writing with Quarto | Data engineering consultations; working session | Technical writing output; weekly summary |
| 6 | 6/15 | PD: companies and networking | Project working session | Weekly summary |
| 7 | 6/22 | Statistics consultations | Project working session | Data summary; weekly summary |
Weekly topics may shift; confirm guest sessions on Canvas as the term evolves.
Weeks 8 to 14 (topics abbreviated)
| Week | Date | Block 1 (6:00 to 7:30 PM) | Block 2 (8:00 to 10:00 PM) | Deliverables due end of class unless noted |
|---|---|---|---|---|
| 8 | 6/29 | PD: behavioral interview prep | Project working session | Weekly summary |
| 9 | 7/6 | Machine learning consultations | Project working session | Weekly summary |
| 10 | 7/13 | PD: mock interviews I | Project working session | Weekly summary; poster rough draft |
| 11 | 7/20 | PD: internal mock interviews (Portland) | PD continued | Weekly summary |
| 12 | 7/27 | Peer feedback working session | Project working session | Weekly summary; write-up rough draft |
| 13 | 8/3 | Presentation prep | Peer practice; working session | Weekly summary |
| 14 | 8/10 | Poster presentations (Salem) | Poster presentations (Salem) | Final write-up due 8/17 |
Practical expectations (carry everywhere)
- Keep Canvas deadlines and README cadence aligned; surprises should show up on the board first.
- Treat meta-project peers as a standing review panel, not extra teammates.
- Make integration explicit across engineering, modeling, visualization, ethics, and communication so the project cannot be read as a thin dashboard or a notebook-only exercise.
References
Sources
- DATA 510 syllabus (Canvas; local mirror: course syllabus HTML).
- Capstone methodology companion page: project methodology.