Skip to main content

Week 8: Dashboard Prototypes for KT

Overview

This is the first app-forward week. You will move from audited EDA outputs (Week 7) to a small dashboard-style Knowledge Translation product that helps a defined audience explore one descriptive question.

The required core pathway is the Quarto starter dashboard: it renders reliably in every Codespace, makes the analysis visible, and is the file you will adapt for Assignment 7. The Shiny exemplar discussed on dash05 is a demonstration only — your instructor will run it in class; you are not required to install Shiny, run the app, or produce A7 deliverables from it.

ImportantRequired / Draft / Optional This Week
  • Required (graded): the nine Assignment 7 files, explained one by one in dash07.
  • Draft (studio work that feeds A7, not graded on its own): the render note, audience sentence, before/after text, testing log, and limitation statement from dash06.
  • Demo-only (not graded, do not build on it): the Shiny exemplar on dash05 — watch it; do not submit A7 work from it.

Tags like [required] and [demo-only] are defined in How To Read Submission Lists.

Objectives

By the end of the week you can:

  • explain the difference between a static report, a Quarto dashboard-style product, and a Shiny app;
  • render the Quarto starter dashboard locally and adapt one audience-facing element;
  • apply the 1-1-1-1-1 rule to translate one Week 7 finding into a dashboard prototype;
  • distinguish text/UI changes (allowed for A7) from analytic changes (not allowed for A7);
  • test a dashboard locally and record what was checked;
  • apply privacy-safe publishing guidance before sharing any health-data product.

Connection

Week 7 established an auditable descriptive foundation. Week 8 asks how a non-technical audience will encounter that evidence in a dashboard. Week 9 turns one dashboard finding into a concise scientific communication product.

Dashboard Case Study: NHANES Equity

The NHANES Health Equity Dashboard is now the foreground case study. Use it as a small knowledge-translation product built on the cached NHANES equity dataset.

  • App code: examples/nhanes-equity/app/app.R
  • Cached data: examples/nhanes-equity/data/nhanes_equity_v6.rds and examples/nhanes-equity/data/nhanes_equity_v6.csv
  • Optional rebuild script: examples/nhanes-equity/scripts/build_nhanes_equity.R

Run the dashboard from the repository root:

shiny::runApp("examples/nhanes-equity/app")

The app is offline-first: it loads the local RDS cache when available. Rebuilding from CDC uses nhanesA, requires internet access, and is optional unless assigned.

Companion materials:

  • Week 10 report template: weeks/week10-reporting-publishing/report-template.qmd
  • Week 11 static dashboard demo: weeks/week11-portfolio-surgery/static-dashboard-ojs.qmd

Reading order

  1. dash01 — What a Dashboard Is, and Three Pathways
  2. dash02 — From One Finding to a Prototype: The 1-1-1-1-1 Rule
  3. dash03 — Starter Dashboard (required core artifact)
  4. dash04 — Audience-Facing Text: What You Can and Cannot Change for A7
  5. dash05 — Testing, Privacy-Safe Publishing, and the Shiny Demo
  6. dash06 — In-Class Studio: Dashboard Adaptation
  7. dash07 — Assignment 7 Walkthrough and Reference

Class plan

  1. Render the Quarto starter dashboard from the repository root.
  2. Choose a stakeholder audience and write one question that audience could answer from the dashboard.
  3. Identify one control, title, caption, help text, or interpretation note to revise.
  4. Improve that audience-facing text without changing analytic logic.
  5. Re-render the dashboard with a different selected_income value and record the test result.
  6. Watch the optional Shiny exemplar demonstration (dash05) — for awareness only, not for A7.

Student output

By the end of class each student has a local render note, an audience sentence, a before/after text change, a testing log entry, and a limitation statement — the studio drafts of every Assignment 7 file.

Definition of done

  • The Quarto starter dashboard renders locally from the cached NHANES data.
  • The audience-facing change improves clarity without changing the analysis.
  • The testing log records how the dashboard was checked in Codespaces.
  • The privacy-safe publishing note avoids row-level data sharing, small-cell claims, and causal overreach.
  • No A7 submission uses the Shiny app as its primary deliverable.

What students leave with

A tested dashboard adaptation note and a decision rule for the term project: use the simplest pathway that serves the audience, with the Quarto pathway as the supported default and any other pathway requiring instructor approval.