What Do *, **, ***, and **** Mean in Statistical Significance? P Value Significance Stars

What Is a P Value?

Before understanding p value significance stars, it's important to understand what a P value represents. A P value measures how likely it is to observe your data (or something more extreme) if the null hypothesis is true.

• Smaller P values indicate stronger evidence against the null hypothesis.
• Larger P values suggest there isn't enough evidence to conclude a significant difference.

The P value does not measure the size or importance of an effect. It only measures how compatible the observed data are with the null hypothesis.

What Are Significance Stars?

Significance stars are shorthand symbols that represent different ranges of P values. These are represented by asterisks (*). Instead of writing the exact P value every time, researchers often use asterisks to indicate the level of statistical significance. These symbols make tables, charts, and scientific figures much easier to read while still conveying the strength of statistical evidence.

For example, instead of displaying:

P = 0.00003

Prism may simply display:

****

This immediately tells us that the result is highly statistically significant.

Standard Meaning of Significance Stars

GraphPad Prism follows the widely accepted convention for reporting statistical significance asterisks.

This is the standard interpretation used in most biomedical and scientific publications.

What Each Symbol Means

Why Do Some Prism Reports Only Show Three Stars?

One of the most common questions among GraphPad Prism users is why some analyses display a maximum of three significance stars (***), while others can display four stars (****). The answer lies in how Prism formats P values, not in how it calculates statistical significance.

GraphPad Prism always computes the exact P value using the selected statistical test. Whether you're performing a t test, ANOVA, nonparametric test, or regression analysis, the underlying calculation is identical regardless of the reporting format you choose. The difference is simply how the result is presented.

P Value Reporting Formats in GraphPad Prism 11

Starting with GraphPad Prism 11, users can choose how P values appear throughout the software.

Comparison of Prism P Value Formats

Consider Three Hypothetical Experiments

Notice that in APA and NEJM formats, all three results appear simply as P < .001, so they all receive three stars. However, in GraphPad Style, Prism can identify that the last two P values are actually smaller than 0.0001, so they receive four stars (****).

Does Three Stars Mean Less Significant?

Researchers often deal with a common misconception that if Prism shows only three stars, the result is somehow less significant than one displaying four stars, but that is not the case.

If you're using APA or NEJM formatting, Prism is simply following the reporting standards of those styles. A result with an exact P value of 0.0000005 will still be displayed as P < .001 and represented by three stars (***). The underlying statistical significance has not changed, only the way it is displayed.

Multiple t-Tests in GraphPad Prism

The Multiple t-Test analysis behaves slightly differently from other analyses. In Prism, the "Significant" column displayed either a single asterisk (*) or remained blank. It never displayed multiple stars. This is an intentional design choice in GraphPad Prism and often raises questions among users.

When you choose the Statistical Significance approach for multiple comparisons, Prism compares each adjusted P value. The column simply indicates whether the adjusted P value is smaller than your selected threshold:

• Yes – The comparison is statistically significant after applying any selected multiple-comparison correction.
• No – The comparison is not statistically significant.

It's important to note that this Significant column does not indicate the degree of significance. Whether the adjusted P value is 0.049, 0.008, or 0.00001, the result will simply be displayed as "Yes" if all of these values are below the specified threshold. Therefore in GraphPad Prism, this column simply reports Yes or No, making it easier to determine whether each comparison is statistically significant.

Why Are Significance Stars Used?

Researchers use p value stars because they simplify data presentation and make scientific figures easier to interpret. Some advantages include:

• Cleaner graphs and tables
• Faster comparison between groups
• Easier identification of statistically significant results
• Consistent reporting across scientific publications
• Improved readability for readers and reviewers

Why Do Journals Prefer Three Stars?

Many scientific journals, including those following APA and New England Journal of Medicine (NEJM) guidelines, recommend reporting very small P values as P < .001 rather than displaying excessive decimal places. This approach keeps tables and figures cleaner and avoids implying an unrealistic level of precision. Since any P value below 0.001 already represents very strong evidence against the null hypothesis, additional decimal places often provide little practical value in published research.

When Should You Use Four Stars?

If you want to highlight extremely small P values, choose either GraphPad Style or Custom Digits. These formats preserve additional precision and allow Prism to display four significance stars (****) for P ≤ 0.0001. This can be particularly useful when preparing technical reports, regulatory submissions, or detailed statistical analyses where reporting the smallest measurable P values is important.

Should You Report Stars or Exact P Values?

While significance stars are excellent for summarizing results in figures, many journals recommend reporting the exact P value whenever possible. For example, instead of reporting only *** you may report:

P = 0.0007 or P < 0.001

Providing exact P values offers greater transparency while still allowing readers to interpret the strength of the statistical evidence.

Best Practices for Reporting Statistical Significance

• Use significance stars consistently throughout your manuscript, and define every symbol in the figure legend.
• For the sake of transparency, always report exact P values whenever journal guidelines recommend it.
• Do not confuse statistical significance with practical or clinical significance.
• Report effect sizes and confidence intervals alongside P values whenever appropriate.