Graphing data
Open the Build a graph that reveals the pattern room
This article opens a matching guided lab. Follow the on-screen prompts, make a prediction, and connect the model result back to this topic.
01 // Setting up the axes
A lab graph almost always puts the independent variable on the horizontal x-axis and the dependent variable on the vertical y-axis, because the graph is showing how the measured value responds to the variable you changed. Each axis needs a label with units, and the graph itself needs a title that names both variables.
Choose a scale that uses most of the graph paper without cramming the data into a corner. Find the range of your data first (lowest to highest value), then pick a scale where each gridline square represents a round number like 1, 2, 5, or 10 units.
02 // Choose a graph that matches the variable
| Data question | Graph | Connect points? | Example |
|---|---|---|---|
| Compare separate categories | Bar graph | No | Mean absorbency by towel brand |
| Show change across time | Time-series line plot | Usually, if samples are ordered in time | Plant height by day |
| Check association between two numeric variables | Scatter plot | Not automatically | Mass versus volume |
| Model a continuous measured relationship | XY line graph | Only when the model/measurement supports it | Pendulum period versus length |
03 // Line graph or bar graph
- Use a time-series line plot for ordered observations such as time, or a scatter plot for two numeric variables. Connect points only when the sampling or model makes intermediate values meaningful; do not connect unrelated pairs just because both axes have numbers.
- Use a bar graph when the independent variable is in separate categories, like different liquids, different brands, or different plant species — there is no meaningful value between the bars.
- Never connect the dots on a bar-graph-style comparison, and never leave gaps between bars on a true histogram of continuous data — the graph type should match what the x-axis actually represents.
04 // Plot an example from raw values
| Day | Height (cm) | Point to plot | Plotting check |
|---|---|---|---|
| 0 | 2.0 | (0, 2.0) | Start at the labeled origin |
| 2 | 3.1 | (2, 3.1) | Move 2 days right, then 3.1 cm up |
| 4 | 4.0 | (4, 4.0) | Keep intervals equal |
| 6 | 4.8 | (6, 4.8) | Plot measured values, not estimates |
05 // Building the graph
1. Find your range
Look at your lowest and highest data values on each axis before choosing a scale.
2. Pick a round scale
Choose a scale per gridline square (1, 2, 5, or 10) that spreads the data across most of the paper.
3. Label both axes
Write the variable name and unit on each axis, for example "time (minutes)" or "height (cm)."
4. Plot and connect or bar
Plot each data point, then connect them with a line (continuous data) or draw separate bars (categorical data).
06 // Plot points and show variation
Plot each data point carefully at the intersection of its x and y values. Use a small dot or clear symbol so the exact location is visible. If multiple trials were collected, show individual values when possible or explain how the plotted mean was calculated. Add error bars only when you know what they represent, such as standard deviation or measurement uncertainty.
Connect points only when doing so makes sense for the data and question. A line through time-ordered observations can help show change; a best-fit line for a scatterplot summarizes an overall association and need not pass through every point. Do not draw a smooth curve that implies unsupported measurements. Include a legend when more than one group or series appears.
07 // Interpret without overclaiming
Describe the pattern with specific evidence: state whether the measured response rose, fell, stayed similar, or varied, and cite representative values. A graph can suggest an association, but it does not by itself prove that the independent variable caused the change. Confounding variables, small samples, and measurement uncertainty can alter the apparent pattern.
Before drawing a conclusion, compare the graph with the original data table to catch plotting mistakes. Check units, axis orientation, missing values, and whether the scale has compressed important variation. A clear graph supports discussion, but it cannot repair an uncontrolled experiment or replace honest reporting of unusual observations.
08 // Keep individual trials visible
Plot each trial when the graph type allows it instead of showing only an average. Individual points reveal spread and unusual measurements. Label whether a summary is a mean, median, or another statistic.
Keep the original values in a table so readers can check the graph. A neat average should not hide inconsistent measurements or missing trials.
09 // Describe the pattern without extending beyond the data
Check units, axis directions, interval spacing, and every coordinate against the raw table. Keep categories separate and connect points only when order and question support it.
If a point seems unusual, check the original record and document any procedural reason before excluding it. Preserve all trials in the source table.
Connect the next step
Use these related skills to carry the same measurements into your next stage:
10 // Questions people ask
Which variable goes on the x-axis of a lab graph?
The independent variable — the one the experimenter chose and changed on purpose — goes on the horizontal x-axis. The dependent variable, the one measured as a result, goes on the vertical y-axis.
When should you use a bar graph instead of a line graph?
Use a bar graph when the independent variable falls into separate categories with no numeric value between them, like different liquids or brands. Use a line graph when the independent variable is continuous, like time or temperature.
How do you choose a scale for a graph?
Find the lowest and highest values in your data, then pick a round number per gridline square (such as 1, 2, 5, or 10) that spreads your data across most of the paper without cramming it into a small corner.
