Home/Chart Types/Correlation/Connected scatter plot
CorrelationIntermediate

Connected Scatter Plot

A scatter plot where consecutive points are connected by lines, tracing how the relationship between two variables evolves over time.

// 01 — The chart

What it looks like

Example — GDP vs. life expectancy over decades1960 – 2020
80706050$5k$20k$50kGDP per capitaLife expectancy196019701980199020002020

A connected scatter plot tracing how GDP per capita and life expectancy co-evolved from 1960 to 2020. Year labels mark each time step.

// 02 — Definition

What is a connected scatter plot?

A connected scatter plot is a hybrid between a scatter plot and a line chart. Like a scatter plot, it maps two variables to the X and Y axes. Like a line chart, it connects the points in sequence — typically chronological order. The result is a trajectory that shows how the relationship between two variables changes over time.

This chart type excels when you want to answer: “How have these two variables moved together (or apart) over time?” The path traced by the line reveals acceleration, loops, reversals, and phase shifts that neither a scatter plot nor a line chart alone can show.

The connected scatter plot gained modern popularity through Hans Rosling’s animated Gapminder presentations, which traced nations’ GDP and life expectancy trajectories decade by decade.

Key difference from scatter plots: A regular scatter plot treats each point independently; a connected scatter plot adds the dimension of sequence, revealing the path through the data rather than just the cloud of values.

// 03 — Anatomy

Parts of a connected scatter plot

ABCDt₁E
A — Y-axis: One of the two variables being tracked
B — X-axis: The second variable — not time (time is implicit)
C — Data point: Each dot is one observation at a specific time step
D — Connecting line: Segments connecting consecutive time steps, forming the trajectory
E — Time label: Annotations showing which time period each point represents

// 04 — Usage

When to use it — and when not to

✓Use a connected scatter plot when…
  • You want to show how two variables co-evolve over time
  • The temporal trajectory is the story — loops, reversals, or acceleration
  • Comparing a small number of entities (1–3) over many time steps
  • Highlighting phase shifts — when one variable leads or lags the other
  • Presenting economic, demographic, or scientific time-series pairs
  • You want to compress three dimensions (X, Y, time) into a 2D plot
×Avoid a connected scatter plot when…
  • You have many overlapping entities — lines will tangle into spaghetti
  • The time order doesn’t matter — a regular scatter plot is clearer
  • Your audience isn’t familiar with the format — it’s less intuitive than line charts
  • You have very few time steps (2–3) — a paired bar chart might be simpler
  • The data is noisy and has many crossings, making the path hard to follow
  • You want to show a single variable over time — use a standard line chart

// 05 — Reading guide

How to read a connected scatter plot

Follow these steps whenever you encounter a connected scatter plot.

1

Identify the two axis variables

Unlike a line chart where X is time, both axes here represent data variables. Time is encoded implicitly in the sequence of connected points.

2

Find the starting point

Look for a label (usually a year or date) marking the first observation. This is where the trajectory begins.

3

Follow the path chronologically

Trace the line from the start, noting the direction. Movement to the upper-right means both variables are increasing; movement down-left means both are decreasing.

4

Look for loops and reversals

If the path doubles back on itself, it means the relationship reversed — one or both variables changed direction. These inflection points are often the most interesting part of the story.

5

Compare segment lengths

Longer segments mean faster change between consecutive time steps. Short, clustered segments indicate stability or stagnation.

// 06 — Data format

What your data should look like

Each row represents one time step with at least three columns: the time variable, the X-axis variable, and the Y-axis variable. Rows must be ordered chronologically.

// Time-series pair

| year | gdp_per_capita | life_expectancy |

|------|----------------|-----------------|

| 1960 | 3,200          | 52.4           |

| 1980 | 12,400         | 65.1           |

| 2020 | 48,700         | 78.9           |

// 07 — Construction

How to build one

1.

Sort your data by time — the connecting lines must follow chronological order to make sense.

2.

Map the first variable to the X-axis and the second to the Y-axis, choosing appropriate scales.

3.

Plot each observation as a dot at the (X, Y) coordinate for its time step.

4.

Connect consecutive dots with line segments in temporal order.

5.

Add time labels — annotate points with year/date labels so readers can follow the chronology. At minimum label the start and end points.

// 08 — Pitfalls

Common mistakes

Omitting time labels

Without year or date annotations, the reader has no way to know when each point occurred. The temporal context is the entire purpose of this chart type.

Too many overlapping paths

Plotting more than 2–3 entities on the same chart creates visual spaghetti. Use small multiples instead for many comparisons.

Unequal time intervals without indication

If your data jumps from 2005 to 2020, the connecting line implies smooth change. Mark or annotate gaps in the time series.

Misleading axis scales

Different axis scales can exaggerate movement in one direction. Ensure both axes are scaled appropriately to the data range.

Using it when time order doesn’t matter

If there’s no meaningful sequence to the data, the connecting lines add confusion rather than insight. Use a regular scatter plot.

// 09 — In the wild

Real-world examples

Gapminder

Hans Rosling’s famous trajectories showing how countries moved through GDP-vs-life-expectancy space over decades, revealing dramatic improvements in health and wealth.

Climate science

Plotting CO₂ concentrations against global temperature anomalies over time to show the accelerating divergence since the Industrial Revolution.

Sports analytics

Tracking a team’s offensive and defensive ratings season by season to visualize whether they improved in both dimensions or traded one for the other.

// 10 — At a glance

Quick reference

Also known as

Trajectory plot, temporal scatter plot

Category

Correlation

Typical data

Two numeric variables + time

Best for

Showing co-evolution of two variables

Difficulty

Intermediate

Requires

Sequentially ordered observations

// 11 — Accessibility

Making it accessible

Label each data point with its time value to make the path readable without relying on color

Use distinct line styles (dashed, dotted) when comparing multiple trajectories

Provide a data table alternative showing time, X, and Y values for screen readers

Add directional arrows to show which way time flows along the path

Use large enough point markers and sufficient spacing for low-vision readers

// 12 — Variations

Common variations

Animated connected scatter

Points appear sequentially in an animation, letting the viewer watch the trajectory unfold in real time.

Multi-entity trajectories

Multiple colored paths on the same axes — best with 2–3 entities using distinct line styles.

Arrow-headed path

Arrowheads on each segment indicate time direction, eliminating the need for year labels on every point.

Bubble-connected scatter

Point size encodes a third variable (e.g., population) at each time step, adding another data dimension.

// 13 — FAQs

Frequently asked questions

What is a connected scatter plot?+

A connected scatter plot is a hybrid between a scatter plot and a line chart. Like a scatter plot, it maps two variables to the X and Y axes. Like a line chart, it connects the points in sequence — typically chronological order. The result is a trajectory that shows how the relationship between two variables changes over time.

When should you use a connected scatter plot?+

Use a connected scatter plot when you want to show how two variables co-evolve over time. It also works well when the temporal trajectory is the story — loops, reversals, or acceleration, and when comparing a small number of entities (1–3) over many time steps.

When should you avoid a connected scatter plot?+

Avoid a connected scatter plot when you have many overlapping entities — lines will tangle into spaghetti. It is also a poor fit when the time order doesn’t matter — a regular scatter plot is clearer, or when your audience isn’t familiar with the format — it’s less intuitive than line charts.

What data do you need to make a connected scatter plot?+

Each row represents one time step with at least three columns: the time variable, the X-axis variable, and the Y-axis variable. Rows must be ordered chronologically.

How is a connected scatter plot different from a scatter plot?+

Both a connected scatter plot and a scatter plot can look similar at first glance, but they answer different questions. Reach for a connected scatter plot when the comparisons and patterns it was designed to reveal match what you need to communicate, and choose a scatter plot when its particular strengths better fit your data and audience.

What is another name for a connected scatter plot?+

Connected Scatter Plot is also known as Trajectory plot, temporal scatter plot. The name varies between fields, but the visualisation technique is the same.

What size of dataset works best for a connected scatter plot?+

Connected Scatter Plot works best for Showing co-evolution of two variables. Outside that range the chart either looks empty or becomes too cluttered to read clearly.

Are connected scatter plots accessible to screen readers?+

Yes — a connected scatter plot can be made accessible to screen readers by pairing it with a clear text summary of the key insight, ensuring color choices meet WCAG contrast guidelines, adding descriptive alt text or aria-label to the SVG, and offering the underlying data as an HTML table fallback for assistive technologies.