Skip to content

brycewang-stanford/Auto-Empirical-Research-Skills / skills/14-luischanci-claude-code-research-starter/dot-claude/skills/data-analysis

data-analysis

Verified

End-to-end R data analysis workflow from exploration through regression to publication-ready tables and figures.

Data Analysis

Run an end-to-end data analysis in R through 5 phases.

Phase 1: Setup and Data Loading

  • Read R code conventions from .claude/rules/r-code-conventions.md
  • Create script with proper header (title, author, purpose, inputs, outputs)
  • Load packages with library() (never require())
  • Set set.seed() if stochastic elements are involved
  • Load data using relative paths

Phase 2: Exploratory Data Analysis

  • Summary statistics (mean, sd, min, max, N)
  • Missingness patterns
  • Distributions of key variables
  • Bivariate relationships
  • Time patterns (if panel data)
  • Group comparisons
  • Save diagnostics to output/diagnostics/

Phase 3: Main Analysis

  • Use fixest for panel data with fixed effects (see /econometrics-r for detailed patterns)
  • Use lm/glm for cross-section
  • Cluster standard errors appropriately
  • Run multiple specifications (simple to complex)
  • Report both statistical and economic significance
  • For causal inference (IV, DiD, RDD), see /causal-inference-r for diagnostics and modern estimators
  • For spatial econometrics, see /econometrics-r for spdep/spatialreg patterns

Phase 4: Publication-Ready Output

  • Tables: modelsummary (preferred) or stargazer, export as .tex and .html
  • Figures: ggplot2 with project theme, explicit dimensions, white background, 300 DPI
  • Export figures as .pdf and .png

Phase 5: Save and Review

  • saveRDS() for all computed objects
  • Run r-reviewer agent on generated script
  • Address Critical/High issues

Script Structure Template

# ==============================================================================
# Title: [Analysis Name]
# Author: [Name]
# Date: [YYYY-MM-DD]
# Purpose: [What this script does]
# Inputs: [Data files]
# Outputs: [Tables, figures, RDS files]
# ==============================================================================

# 0. Setup ----
library(tidyverse)
library(fixest)
library(modelsummary)

# 1. Data Loading ----

# 2. Exploratory Analysis ----

# 3. Main Analysis ----

# 4. Tables and Figures ----

# 5. Export ----

All scripts saved to scripts/R/, all outputs to output/.

planning-and-task-breakdownaddyosmani/agent-skillsBreaks work into ordered tasks. Use when you have a spec or clear requirements and need to break work into implementable tasks. Use when a task feels too large to start, when you need to estimate scope, or when parallel work is possible.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
screenpipethedotmack/claude-memSearch screen and audio history recorded by Screenpipe and optionally save selected context into claude-mem. Use when the user asks to recall desktop activity, a meeting, or something seen in another app through Screenpipe.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
contact-widgetnexu-io/open-designSelf-contained floating chat widget with welcome screen, social links, meeting button, and message input. Single HTML file, zero dependencies.Stars: 108kUpdated Claude CodeCodexVerifiedInstall
meeting-notesnexu-io/open-designMeeting notes page — title bar with attendees, agenda checklist, decisions block, action items table with owners + dates, and a "next meeting" footer. Use when the brief mentions "meeting notes", "minutes", "1:1 notes", "all-hands recap", or "会议纪要".Stars: 108kUpdated Claude CodeCodexVerifiedInstall