Scatter Plot Generator – See the Relationship in Your Data
The Scatter Plot Generator plots one point per observation from two numeric variables, making it the go-to chart for questions like do these move together? Hours studied against exam scores, advertising spend against revenue, temperature against ice-cream sales — paste the pairs and the pattern (or lack of one) appears instantly. The tool runs entirely in your browser, supports up to eight comparison groups, and fits a least-squares trend line with the equation and R² shown beneath the chart.
From Rows of Numbers to Points
Each table row is one observation: the X value first, then a Y value per series. Edit cells directly, copy columns straight from Excel or Google Sheets into the Paste tab, or upload a CSV file — a header row is detected automatically and becomes the series names:
Hours,Class A,Class B
1,52,48
2,58,50
3,65,59Rows whose X value isn't a number are skipped (the tool tells you how many), and a missing Y cell simply omits that point for that series — nothing silently becomes a zero.
Trend Lines You Can Quote
Turn on Show trend line and every series with at least two points gets an ordinary least-squares regression line drawn through its data, dashed so it never masquerades as a measurement. Below the chart, each series reports its fitted equation in y = mx + b form, the R² goodness-of-fit (how much of the variation the line explains, from 0 to 1), and the number of points used. That's everything you need for a lab report or a quick sanity check on a claimed correlation.
Compare Groups on One Plot
Add a series column per group — control vs. treatment, this year vs. last year, region A vs. region B. Each group gets its own color from a colorblind-safe palette applied in a fixed, maximally distinguishable order, its own trend line, and its own statistics row. A legend appears automatically, and hovering near any point shows its exact X and Y in a tooltip. Use the point size slider to make dense clouds readable or sparse data more prominent.
Reading a Scatter Plot Well
Look for three things: direction (uphill = positive relationship, downhill = negative), strength (tightly clustered around the trend = strong, diffuse cloud = weak), and outliers (points far from the pattern often deserve their own investigation). And remember the golden rule: even a perfect-looking line shows correlation, not causation — both variables may be driven by something you haven't plotted.
Export-Ready Output
Download the framed chart as a crisp 2×-resolution PNG or a scalable SVG, copy the image to your clipboard for a chat or email, or export the data table as CSV. Light and dark mode are both supported, with colors tuned separately for each background.