How to Write a Figure Caption for a Lab Report

A strong figure caption lets a reader understand the visual without hunting through the lab report. Start with the figure number and a precise title, then state what was measured, identify the groups or conditions, explain symbols or error bars, and include only the method details needed to interpret the image. Put the caption below the figure unless your course guide says otherwise. This article is for undergraduate science and engineering students who need a repeatable way to write captions for graphs, photographs, diagrams, and multi-panel figures.

Key Takeaways

• Make the caption self-contained: a reader should know what the figure shows, which conditions were compared, and how the data are represented.
• Use the sequence label, message, setup, encoding, and scope. This five-part formula prevents both vague one-line captions and miniature methods sections.
• Name every symbol, line, panel, abbreviation, and error-bar type that the reader cannot decode from the axes alone.
• Follow the assignment or disciplinary style guide when it conflicts with general advice; caption conventions vary across fields.

What a Lab Report Figure Caption Must Do

A caption is not merely “Figure 1” plus a topic. The University of North Carolina Writing Center describes figure captions as concise but comprehensive: they identify the data, direct attention to important features, and can sometimes include interpretation. That standard gives you a useful test. If a classmate can explain the visual after reading only the figure and caption, the caption is doing its job.

The caption and the figure divide the work. Axis labels carry variable names and units. A legend maps colors, shapes, or line styles to groups. The caption supplies the scientific context that those labels cannot: the comparison, treatment, sample, summary measure, uncertainty display, and any detail essential to interpretation. It should complement the Results paragraph rather than duplicate it word for word.

The Cornell Engineering communications guide separates a caption into a label, title, informative explanation, and a citation when the visual or data came from another source. Cornell also encourages interpretive captions in technical documents. Some laboratory courses are stricter and reserve interpretation for the Results or Discussion, so treat your rubric as the final authority.

Use the Five-Part Caption Formula

Draft the caption in this order. You can combine parts into two or three sentences, but checking them separately makes omissions easy to spot.

1. Label and number. Write “Figure 1.” or the abbreviation required by your course. Number figures in the order they are first discussed, independently from tables.
2. Message. Give the visual a specific title or opening statement. “Effect of salt concentration on seedling growth” tells the reader more than “Plant results.”
3. Experimental setup. Identify the organism, material, treatment, comparison, and time point needed to understand the visual. Include temperature, concentration, or sample size only when it changes the meaning of the display.
4. Visual encoding. Define bars, lines, symbols, panels, abbreviations, and error bars. If points show means and whiskers show standard deviations, say so explicitly.
5. Scope or takeaway. State what relationship or result the reader should notice if your discipline permits interpretation in captions. Do not make a causal claim that the experiment cannot support.

💡 A compact drafting template: Figure X. [Specific message]. [What was measured] for [groups or conditions] after [relevant procedure or time]. [How values and uncertainty are represented]. [What symbols or panels mean].

Build a Figure Caption Step by Step

Consider a fictional biology lab in which students measured root length after exposing seedlings to three salt concentrations. The values below are invented only to demonstrate caption structure; they are not research findings.

The weak version

Figure 1. Results of the salt experiment.

This caption has a number, but it does not identify the response variable, treatment levels, duration, organism, sample size, summary statistic, or error bars. A reader must search the Methods section before the graph makes sense.

Add the scientific message and setup

Begin with the relationship: “Salt concentration and radish seedling root length after 48 hours.” Then add the conditions: seedlings were exposed to 0, 50, or 100 millimoles per liter sodium chloride. The title now tells the reader what changed, what was measured, and when the measurement occurred.

Explain how the values are represented

Suppose each bar is the mean of eight seedlings and each error bar is one standard deviation. Those details belong in the caption because the graph alone cannot reveal them reliably. If letters above the bars mark groups that differ in a statistical comparison, the caption must also define the letters and name the test used. Do not write only “error bars shown.”

The finished caption

Figure 1. Salt concentration and radish seedling root length after 48 hours. Seedlings were exposed to 0, 50, or 100 millimoles per liter sodium chloride. Bars show mean root length for eight seedlings per condition, and error bars show one standard deviation. Fictional data are used for this caption example.

The finished version is self-contained without repeating every procedural step. It leaves equipment settings and the full protocol in Methods, while preserving the information needed to read the plotted values.

A simple workflow helps when your lab notes are scattered. After the experiment, extract the treatment, response variable, units, sample size, and uncertainty measure into one note. Snitchnotes can turn that note into a short checklist or practice question set, making it easier to verify the caption before submission.

Adapt the Formula to Different Figure Types

Graphs and plots

For a graph, identify the independent and dependent variables, groups, summary statistic, and uncertainty. The Utah State University Engineering Writing Center specifically advises naming the error-bar meaning, such as a standard deviation, standard error, or 95 percent confidence interval. Also define fitted lines or statistical markers if they are not obvious from the legend.

Photographs and micrographs

For an image, name the specimen or material, treatment, imaging method when relevant, and the feature the reader should inspect. Explain arrows, labels, stains, or false color. Include a scale bar in the image and identify it in the caption when size matters; magnification alone may not remain accurate if an image is resized. Avoid claiming that a representative photograph proves a pattern across all samples.

Diagrams and apparatus schematics

For a diagram, state what process or setup is represented, then define arrows, stages, components, and flow direction. If you adapted the diagram from another source, cite that source in the caption. Distinguish a conceptual diagram from a drawing made to scale so the reader does not infer precision that is not present.

Multi-panel figures

Introduce the overall question first, then explain panels in order as A, B, C, and so on. Use one sentence per panel when the content is complex. Define any colors or symbols used across panels once, and state whether panels share a sample, scale, or analysis. Each panel should be mentioned; unexplained panels make the figure incomplete.

Common Figure Caption Mistakes

• Repeating the axis labels. “Temperature versus rate” adds no context. Name the system, treatment, and purpose of the comparison.
• Writing a miniature Methods section. Keep only the procedural details required to interpret the figure. Put reproducible step-by-step methods elsewhere.
• Leaving uncertainty undefined. Standard deviation, standard error, and confidence intervals do not mean the same thing. Name the exact measure.
• Using vague references. Replace “the samples,” “various conditions,” and “results” with the actual groups, values, or variables.
• Reporting unsupported causation. If the design establishes an association, write “was associated with” or describe the observed difference instead of saying one variable caused another.
• Forgetting source credit. Cite visuals or data reproduced or adapted from elsewhere, using the citation style required for the report.
• Mixing table and figure conventions. Figure captions usually sit below figures, while table titles usually sit above tables. Keep their numbering sequences separate.

A Two-Minute Figure Caption Checklist

Read the figure and caption without the surrounding report. Then check each item:

• The figure has the correct number and a specific descriptive title.
• The treatment, comparison groups, response variable, and relevant time point are named.
• Units appear on axes or are explained where necessary.
• All lines, colors, symbols, abbreviations, and panels are defined.
• The caption identifies the summary statistic and the exact error-bar measure.
• The sample size is included when it is needed to interpret the data or required by the course.
• The wording distinguishes observed results from explanations or causal claims.
• Any borrowed or adapted visual or dataset has a citation.
• The caption can be understood without searching the Methods section.
• Formatting matches the rubric and is consistent with every other caption in the report.

When Caption Rules Disagree

General guides do not always make the same editorial choice. For example, the Reed College laboratory report guide asks students to include method context, data expression, sample size, and abbreviations in figure legends, while keeping Results and Discussion out of them. Cornell presents more interpretive technical captions. Both can be sensible within their own contexts.

Use this priority order: assignment rubric, instructor directions, departmental guide, required citation style, then general writing advice. If the instructions remain ambiguous, show the instructor one draft caption and ask whether interpretation belongs in the caption or only in Results. That single question resolves the most common disciplinary difference.

Frequently Asked Questions

How long should a lab report figure caption be?

A caption should be long enough to make the figure understandable on its own and no longer. A simple graph may need two sentences; a multi-panel scientific figure may need several. Include the comparison, context, data representation, and symbol definitions, then remove procedural details that are not necessary for interpretation.

Should a figure caption go above or below the figure?

In most science and engineering lab reports, a figure caption goes below the figure, while a table title goes above the table. However, course and journal styles can differ. Follow the placement specified in the assignment, department handbook, or required style guide, and use the same placement throughout the report.

Do I include results in a figure caption?

Include enough description to direct attention to the displayed pattern, but check the course convention before interpreting it. Some technical-writing guides welcome a concise takeaway; some lab manuals keep interpretation in the Results or Discussion section. Never add a conclusion that the data or experimental design cannot support.

What should I say about error bars?

Name exactly what the error bars represent, such as one standard deviation, standard error, or a 95 percent confidence interval. Also state the sample size when required or useful. The phrase “error bars are shown” is incomplete because readers cannot infer which uncertainty measure you calculated.

Should methods appear in the caption?

Include only methods needed to understand the figure: treatment level, duration, temperature, imaging method, or another condition that changes interpretation. Leave the complete protocol in the Methods section. The caption should make the visual readable, not make the experiment reproducible by itself.

Conclusion

To write a figure caption for a lab report, give the reader a numbered, self-contained explanation of what the visual shows and how to decode it. Use the five-part formula: label, message, setup, visual encoding, and scope. Then test the caption away from the main text and revise every vague noun, undefined symbol, and unexplained error bar.

Before submitting, compare the result with your rubric and run the two-minute checklist. If you use Snitchnotes to organize your laboratory notes, keep the caption fields beside the experiment summary so the treatment, units, sample size, and uncertainty measure are ready when you build the final figure.

Sources

• University of North Carolina Writing Center — Figures and Charts
• Reed College Online Writing Lab — Laboratory Report Instructions
• Utah State University Engineering Writing Center — Tables and Figures
• Cornell Engineering Communications — Captions for Figures in Documents