Scatter plots and line charts both place points on an X/Y grid. They look similar. But they represent different data relationships and answer completely different questions. Choosing the wrong one misleads your audience.
If your goal is to test the relationship between two numeric variables, start with the free scatter plot maker.
Key difference
| Scatter plot | Line chart | |
|---|---|---|
| X axis type | Any numeric value | Time or ordered sequence |
| Connects the dots | No | Yes |
| Purpose | Show relationship between two variables | Show change over one variable (usually time) |
| Each dot represents | One observation (two measurements) | One time point (one measurement) |
| Order of dots | No meaningful order | Sequential — order matters |
When to use a scatter plot
A scatter plot is for two numerical variables measured on the same subject. The question is: does X tell us anything about Y?
- Height vs weight for a group of people
- Study hours vs exam scores
- Temperature vs ice cream sales
- Ad spend vs revenue
- Any "does X correlate with Y?" question
The X axis is a measurement, not a sequence. The dots are independent observations — swapping their order doesn't change the story.
When to use a line chart
A line chart is for one variable measured repeatedly over time (or another ordered sequence). The question is: how did this change?
- Monthly revenue over 12 months
- Daily temperature throughout a week
- Stock price over one year
- Website traffic by day
- Any "how did this change over time?" question
The line connecting dots implies continuity — the value is assumed to change gradually between measured points. This only makes sense when X is time or an ordered sequence.
When to connect the dots
The most important difference is whether to connect dots with lines:
Connect dots when X is time or a sequence, and you're tracking one thing changing. The line communicates: "this is a continuous trend."
Don't connect dots when X is a measurement and dots are independent observations. Connecting them implies an order that doesn't exist — if you swap two data points on a scatter plot, nothing changes. If you swap two time points on a line chart, the whole story changes.
A common mistake: time on X, but scatter plot intended
When someone plots sales in January and February as two dots and draws a line between them, they're making a line chart — even if they call it a scatter plot. If the story is "sales went up from January to February," the line is correct. If the story is "do months with higher advertising have higher sales?", connecting the dots implies a trend between the months themselves, which is misleading.
Same data, two different stories
Imagine you have data on 12 months: average temperature and ice cream sales for each month.
- Scatter plot — temperature (X) vs sales (Y). Each month is one dot. You're asking: does warmer weather correlate with more ice cream sales? The months aren't connected.
- Line chart — month (X) vs sales (Y). You're asking: how did sales change through the year? The months are connected in sequence.
Both use the same data. Both are correct. But they answer different questions.
Scatter plot vs line chart for trends
A scatter plot with a trendline shows a statistical relationship — the overall direction derived from the entire dataset. A line chart shows actual change over time — connecting specific measured values in order.
A rising trendline on a scatter plot means "on average, higher X is associated with higher Y." A rising line on a line chart means "Y increased over time."
FAQ
Can a scatter plot have time on the X axis?
Yes — and then it looks like a line chart without the connecting lines. This is appropriate when you want to show multiple series of dots over time without implying a smooth trend between points, or when you're showing the scatter of individual measurements at each time point.
Should I connect the dots in my scatter plot?
Only if the X axis is ordered (time, sequence, ranked categories) and connecting implies a meaningful trend. For two measurement variables, leave the dots unconnected.
Can I show a trendline on a scatter plot?
Yes — a linear trendline (regression line) is a standard addition to scatter plots. It shows the overall statistical relationship. This is different from connecting individual dots; the trendline is calculated mathematically from all the data.
What if I have both a time variable and two measurement variables?
You might use a scatter plot with color or size encoding the time dimension, or create separate scatter plots for different time periods. Alternatively, use a line chart per measurement variable. The right choice depends on which relationship you're emphasizing.
How this guide is reviewed
This guide is maintained by the ChartsMakers Editorial Team and reviewed for statistical accuracy, product behavior, and clarity. When a guide relies on software behavior or statistical definitions, we check it against official documentation or established references. See our editorial policy.