馃帗 Journalism Training
Courses

Data, privacy and safety

馃搳 Data Journalism and Basic Statistics

Find public data, build and clean a spreadsheet from scratch, and do the handful of calculations that most local stories need. Then read uncertainty honestly, avoid the classic statistical traps, and publish charts and numbers a skeptic can check.

9 lessons 路 about 2.5 hours 路

Start lesson 1

  1. 1Finding the Data Before You Need ItKnow which federal, state and local agencies publish which numbers, how to read an American Community Survey margin of error, and how to ask for a database through a records request.16 min
  2. 2Spreadsheets From ZeroRows, columns, sorting, filtering and the formulas a reporter uses every week, worked on a small invented county checkbook.18 min
  3. 3Pivot Tables: Answering Who, How Much and How OftenUse COUNTIFS, SUMIFS and pivot tables to group thousands of rows by vendor, department or month in a few clicks.15 min
  4. 4Cleaning Data: The Unglamorous 80 PercentFix inconsistent names, broken dates, numbers stored as text, duplicate rows and stray totals, and pull tables out of PDFs with Tabula.17 min
  5. 5The Statistics That Matter in Local StoriesMean versus median, percent versus percentage points, rates per 1,000 or 100,000, inflation adjustment, small-number traps and the average of averages.18 min
  6. 6Uncertainty: Samples, Margins of Error and SignificanceRead polls and surveys honestly: what a margin of error covers, why a lead needs about twice the margin, what confidence intervals and statistical significance mean, and what they do not.18 min
  7. 7Correlation, Causation and the Traps Between ThemTell correlation from causation, find the confounder, recognize regression to the mean, and work through Simpson's paradox with a real table.17 min
  8. 8Charts and Maps That Tell the TruthPick the chart that fits the data, set honest axes, map rates instead of counts, and build it free in Datawrapper.16 min
  9. 9Checking Your Numbers and Writing Them UpAudit every number before publication, round honestly, keep paragraphs readable, show your method and publish the data so readers can check you.16 min