Example: M1 — Project Proposal

Illustrative example — do not copy

This is one worked example on the class NHANES dataset, written by a fictional group, to show the expected structure and depth of a good submission. Your project must use your own question, data, audience, and analysis — do not copy this text or these files. This example lives in the public book for learning; graders assess your original work.

Below is the Cedar Equity Lab group’s m1-proposal submission from the worked NHANES example. See the M1 brief for what is required.

File: ai-use-note.md

AI-use note

AI helped shorten the audience and KT-purpose paragraph. The group checked the dataset dimensions in R, verified the source and income-group derivation in the course files, and removed language that implied causality or national representation. The questions and feasibility decision were approved by all three group members.

File: data-intake-card.md

Data intake card

Dataset: NHANES Health Equity classroom dataset
Source and steward: CDC/National Center for Health Statistics, National Health and Nutrition Examination Survey
Access: Prepared CSV committed at data/nhanes_equity_v6.csv; source documentation at https://wwwn.cdc.gov/nchs/nhanes/; checked 2026-10-01
Terms: Public-use NHANES files; cite CDC/NCHS documentation
Geography and time: United States; cycles represented from 1999-2000 through 2017-2018 and 2021-2023
Format and level: CSV; row-level public-use-derived classroom records
Unit of analysis: One examination participant record within a survey cycle
Key variables: Age in years; BMI in kg/m²; IncomeGroup derived from the family income-to-poverty ratio; Cycle as the NHANES cycle label
Privacy/security: Use only aggregate outputs, avoid small-cell claims, and do not upload row-level records to external AI tools
Stewardship: Describe group comparisons without stigma and keep the prepared, unweighted classroom boundary beside results
Indigenous Data Sovereignty relevance: No specific Indigenous community, service, territory, or identifier is part of the proposed comparison. We will reassess CARE/OCAP relevance if the project scope changes.
Known limitations: Population inference requires the correct NHANES survey design; the prepared file is not a causal dataset
KT framing: The data can identify a descriptive pattern worth discussing while teaching the audience what the comparison cannot establish
Citation: CDC/NCHS, National Health and Nutrition Examination Survey, https://wwwn.cdc.gov/nchs/nhanes/, accessed 2026-10-01

AI could misstate the IncomeGroup cut-points or describe the group as an individual income measure. We checked the variable creation in the course case-study build script and will describe it as a derived category.

File: initial-visual-plan.md

Initial visual plan

Planned figure: a line plot of unweighted mean BMI across NHANES cycles, with one line for each income group and visible cycle and BMI labels.
Audience purpose: help briefing staff see whether the small overall income-group difference is similar across cycles or concentrated in a few periods.
Misinterpretation risk: connected lines may be read as a causal trend or a national estimate. The title and caption will say that the summary is descriptive, unweighted, and based on the prepared classroom records.

File: proposal.md

Project proposal

Project and team

Title: Income and BMI Patterns in NHANES Adults

Group: Cedar Equity Lab — Maya Chen, Jordan Singh, and Alex Martin

Audience and KT purpose

Our audience is municipal public-health staff preparing an internal health-equity briefing. The project will give them a simple descriptive view of income-group differences in the prepared dataset, while keeping the unweighted and non-causal limits visible. It is intended to support a discussion about what deserves closer survey-aware analysis, not an immediate policy decision.

Questions

Primary question: How does mean BMI differ across income groups among adults age 20-80 in the prepared NHANES Health Equity classroom dataset?

Secondary question: How does mean BMI vary across available NHANES cycles for one selected income group?

Data and planned output

We will use the cached CSV derived from CDC/NCHS NHANES public-use files. The planned outputs are a Table 1-style summary, one cycle-by-income figure, a short Quarto report, and a small Quarto dashboard-style page with one editable income group.

Known limitations

The cached file is a prepared classroom dataset. The proposed summaries are unweighted, will exclude some missing values after reporting them, and cannot support causal or national-population claims. A full NHANES analysis would require survey weights, strata, and primary sampling units.

File: reproducibility-plan.md

Reproducibility plan

Expected folders are data/, report/, dashboard/, and milestones/. The cached public-use-derived CSV will be committed once under data/ and treated as read-only. R source files will load tidyverse and knitr explicitly and use paths relative to the project root.

Another student should be able to open the course Codespace, start at the repository root, and run quarto render on the named source. Generated tables, figures, and HTML files will be regenerated from code rather than edited by hand. Any added dependency will be recorded before use.

Feasibility check: the CSV loaded successfully in the course environment with 105,626 rows and 18 columns. R is the project language; a Python version is not needed for the proposed questions.