Scatter plot maker with a least-squares trend line

How to use the scatter plot maker

1. Paste your data into Chart data: X values in the first column, Y values after it, such as Hours studied,Score then 1,52. Add more columns for more Y series.
2. Keep Add a linear trend line (least squares) ticked to fit a line, or untick it for points only.
3. Set the Point size (2–14 px) and decide whether to Start both axes at zero.
4. Fill in Chart title, X-axis title and Y-axis title, and choose a Colour theme, Background, Width and Height.
5. Copy or download scatter-plot.svg. Below it, the Points table and the Trend lines table can be copied or saved as CSV.

What it does and when to use it

A scatter plot shows whether two measurements move together. Each row is one dot: hours of study against a test score, temperature against sales, ad spend against sign-ups. A cloud that rises from left to right suggests a positive relationship; a falling cloud, a negative one; a round blob, little linear relationship at all.

The trend line turns the eye’s impression into numbers. The slope says how much Y changes for each extra unit of X, and r² says how closely the dots hug that line. If you only need to show a series over time, the line chart maker is the better choice. To get the mean, spread and quartiles of each column first, run the data through the summary statistics calculator.

How it works

The fit is ordinary least squares, worked out per series with x̄ and ȳ as the means:

slope     = Σ(x − x̄)(y − ȳ) ÷ Σ(x − x̄)²
intercept = ȳ − slope × x̄
r²        = [Σ(x − x̄)(y − ȳ)]² ÷ (Σ(x − x̄)² × Σ(y − ȳ)²)

The dashed line is drawn only between the smallest and largest X in that series, so it never claims to predict beyond your data. Axis ticks use rounded steps of 1, 2 or 5 × 10ⁿ. Dots are slightly see-through, so overlapping points look darker.

Worked examples

Study time and scores (the sample). Ten students. The legend reads Score (y = 5.537x + 45.69, r² = 0.986): each extra hour goes with about 5.5 more points, and the line explains almost all the variation. The Trend lines table gives the full precision: Score | 5.537281 | 45.689693 | 0.9857 | 10.

A perfect line. 1,2, 2,4, 3,6 give y = 2x + 0 with r² = 1.000.

Two series and a bad row. Temperature in the first column, ice cream and hot drink sales after it, with one row whose X is “n/a”. The tool reports 1 row without a numeric X value was skipped. Ice creams rise (slope 6.084233, r² 0.9949) while hot drinks fall (slope -3.819654, r² 0.9866).

Limits and tips

• Only straight-line fits are offered; curved patterns will show a lower r² than they deserve.
• One outlier can pull the line a long way. Check the Points table for typing errors before trusting the slope.
• Values like 1,234.5, $45 and 12% are read as numbers. The SVG can be turned into a PNG with the SVG to PNG converter.