Up Next: A Report Someone Can Use
Look back at what Module 4 taught (line, bar, scatter and histogram charts, choosing the chart for the question, titles and labels that explain, honest axes, charts everyone can read, several charts on one saved page), try a one-page chart summary of your own, and see what Module 5 adds, a report someone else can follow and rebuild.
Journeyman · Python · Updated
After this lesson you can
- Describe what Module 5's reports will let you do with the charts and answers you can now make
Key takeaways
- After Module 4 you can draw the chart that fits a question, give it a title that states the finding, and keep its axes, colors and labels honest.
- A one-page summary of four charts, each with a title that states its finding, puts every one of those skills to work.
- A saved chart still needs someone to explain it, and a page someone drew by hand has to be drawn again when next month's data arrives.
- Module 5 turns an analysis into a notebook that tells its story, a script that rebuilds the report from new data, and tests that prove its numbers.
When this module started, your answers were tables. Now they're charts that someone can read in a few seconds, and charts you can defend: the right kind for the question, a title that says what it shows, an axis that starts where it should, and colors that every reader can tell apart. That's what a manager, a city council or a class presentation needs from an analyst.
This lesson looks back at what you can do, gives you a page of charts to make on your own, and shows what Module 5 adds: a report someone else can follow, and rebuild.
What you can do now
After Module 4, you can:
- draw four kinds of chart with
fig, ax = plt.subplots(): a line chart withax.plot, bars withax.barandax.barh, a scatter plot withax.scatterand a histogram withax.hist, or the same charts straight from pandas withdf.plot(ax=ax); - choose the chart from the question: a line for change over time, bars to compare groups, a scatter plot for two things that may move together, a histogram for a spread, and sorted bars of shares for the parts of a whole;
- write titles that state the finding, label each axis with its unit, and point at the one thing that matters with
ax.annotate; - keep the axes honest: bars that start at zero, a zoomed line that says so, and side-by-side charts that share a scale with
sharey=True; - make charts everyone can read, with markers, line styles and labels so color never carries the meaning alone, and alt text for a chart in a report;
- put several charts on one page with
plt.subplots(2, 2), give the page a title withfig.suptitle, and save it withfig.savefig.
Try it: a page for the city council
The best way to find out how much you can do is to make a page with no step-by-step instructions. Use pickups.csv from Exercise: Fix the Misleading Charts (its seed.py makes it), and make a 2×2 page that answers these four questions, one chart each:
- How did each neighborhood's share of households recycling change from 2024 to 2025? Two bars per neighborhood, or one bar of the change.
- Which neighborhood sends the most trash per household? Pounds per household in 2025, sorted.
- Do the months with more trash also have more recycling? One dot per month.
- How much do the neighborhoods' monthly tons of trash vary? The spread of all 144 values.
Give every chart a title that states what it shows, label each axis with its unit, start every bar chart at zero, and save the page as a PNG. Then write one sentence of alt text for it. There's no test suite for this one: when a friend can read each chart's answer without asking you a question, it works.
Where a saved page hits a wall
A page of honest charts is a big step, and two things still limit it:
- A chart needs its story. A chart shows what happened; a reader also wants to know the question, the short answer, how you got it and what the data can't tell them. A picture can't hold all of that.
- Next month means doing it all again. A page you drew by running cells by hand has to be drawn by hand again when new data arrives, and every step is a chance to slip.
What Module 5 unlocks
Module 5 finishes an analysis so someone else can use it:
- A notebook that tells a story: the question, the short answer first, the evidence as tables and charts, and what the data can't show, shared as a web page.
- From notebook to script: the cells you run again and again become functions in a module, and one command rebuilds the whole report, charts included, when a new week's file arrives.
- Testing your analysis: tests that prove each number and each chart is right, the same kind of tests that have checked every exercise in this course.
The charts you've learned to make are what those reports are built from: Module 5 makes them part of something someone can read, trust and run again.
Check yourself
2 questions to try on the lesson page.
So what
Module 4 made your answers visible and honest; Module 5 turns them into a report that explains itself and rebuilds itself from new data.