How to Report Z-Test Results in Academic Papers
By William DavisFebruary 16, 2026
Writing up your z-test results doesn't have to be complicated. Whether you're submitting to a journal, defending your thesis, or finishing a class paper, knowing how to present your statistical findings clearly makes all the difference.
Let's walk through exactly how to report z-test results so your readers understand what you found and why it matters.
Why Proper Reporting Matters
You ran your analysis, got your numbers, and found something interesting. Great! But if you can't communicate those findings clearly, your hard work won't have the impact it deserves.
Academic journals have standards. Your professors have expectations. And your readers need to understand your methods and results to trust your conclusions. Proper statistical reporting isn't just about following rules. It's about being transparent and helping others evaluate your work.
Understanding What You Need to Report
Before we get into the specifics, here's what you'll always need to include when reporting z-test results:
Essential Components:
- The test statistic (your z-value)
- The p-value
- The sample size
- Your conclusion about statistical significance
- Descriptive statistics like means and standard deviations
Think of it like telling a complete story. Your readers need enough information to understand what you tested, how you tested it, and what you found.
Basic Z-Test Reporting Format
Here's the standard way to report a z-test result in text:
"A z-test revealed that [group/condition] (M = X.XX, SD = X.XX, n = XX) was significantly different from [comparison value/group] (M = X.XX, SD = X.XX, n = XX), z = X.XX, p = .XXX."
Let's break this down with a real example. Say you're testing if students who used a new study app scored differently than the national average of 75 on a standardized test.
"A one-sample z-test showed that students using the study app (M = 82.5, SD = 8.3, n = 150) scored significantly higher than the national average of 75, z = 11.06, p < .001."
See how that works? You've told readers the mean score, standard deviation, sample size, your z-value, and the p-value. That's everything they need to evaluate your claim.
Reporting Different Types of Z-Tests
One-Sample Z-Test
When you're comparing a sample mean to a known population value:
"Participants' reaction times (M = 245 ms, SD = 32 ms, n = 200) were significantly faster than the established baseline of 275 ms, z = -13.26, p < .001."
Two-Sample Z-Test
When comparing two groups:
"Customer satisfaction ratings differed significantly between the control group (M = 6.8, SD = 1.2, n = 300) and the experimental group (M = 7.4, SD = 1.1, n = 300), z = 6.32, p < .001, suggesting the new customer service protocol improved satisfaction."
One-Tailed vs Two-Tailed Tests
Always specify which you used. Your choice affects how you interpret the p-value.
One-Tailed Test
"A one-tailed z-test confirmed our hypothesis that..."
Two-Tailed Test
"A two-tailed z-test revealed a significant difference..."
Including Effect Size
P-values tell you if an effect exists. Effect sizes tell you if it matters. Don't skip this step.
Cohen's d is the most common effect size for z-tests. Here's how to add it:
"The new training program resulted in significantly higher test scores (M = 88.2, SD = 9.1, n = 120) compared to the traditional program (M = 82.4, SD = 8.8, n = 120), z = 4.69, p < .001, d = 0.65, indicating a medium effect size."
Effect Size Guidelines:
- Small: d = 0.2
- Medium: d = 0.5
- Large: d = 0.8
Confidence Intervals Add Context
Confidence intervals show the range where the true population parameter likely falls. They're becoming more important in academic writing because they provide more information than p-values alone.
"The 95% confidence interval for the difference in means was [3.2, 8.4], z = 5.21, p < .001."
This tells readers that you're 95% confident the true difference falls somewhere between 3.2 and 8.4 points.
What to Include in Your Methods Section
Before you report results, you need to explain what you did. Your methods section should cover:
Sample Information
Who participated? How many? How were they selected?
Variables
What did you measure? How?
Assumptions
Did you check if your data met z-test requirements? Large enough sample? Population standard deviation known?
Analysis Approach
One-tailed or two-tailed? What alpha level (usually .05)?
Example methods paragraph:
"We conducted a two-sample z-test to compare average response times between groups. The population standard deviations were known from previous studies (σ₁ = 30 ms, σ₂ = 28 ms). We used a two-tailed test with α = .05. Sample sizes exceeded 30 participants per group, satisfying the requirements for the central limit theorem."
Creating Tables for Multiple Comparisons
If you're reporting several z-tests, a table works better than paragraphs of text. Here's a clean format:
Table 1: Z-Test Results for Group Comparisons
| Comparison | Group 1 M (SD) | Group 2 M (SD) | z | p | d |
|---|---|---|---|---|---|
| Treatment A vs B | 45.2 (6.1) | 41.8 (5.9) | 3.89 | <.001 | 0.57 |
| Treatment A vs C | 45.2 (6.1) | 43.5 (6.3) | 1.88 | .060 | 0.27 |
Tables keep everything organized and easy to compare.
Common Mistakes to Avoid
Don't write "p = 0.000"
No p-value is exactly zero. Write "p < .001" instead.
Don't say "highly significant"
Something is either statistically significant or it isn't. The p-value magnitude doesn't make findings "more significant."
Don't report p-values with too many decimals
Two or three decimal places is plenty. "p = .034" not "p = 0.0342857."
Don't forget the sign
Is your z-value positive or negative? It matters for interpretation.
Don't confuse statistical and practical significance
A tiny difference can be statistically significant with a large sample but meaningless in practice. That's why effect sizes matter.
APA Style Guidelines
Most academic papers follow APA format. The rules below summarize the official APA Style guidance on reporting statistics. Key points:
- Italicize statistical symbols: M, SD, z, p, n
- Use leading zeros for numbers that can exceed 1.0 (M = 0.85)
- Don't use leading zeros for correlations, p-values, or proportions (p = .032)
- Round to two decimal places for most statistics
- Report exact p-values unless p < .001
Writing Your Results Narrative
Numbers alone don't tell the story. Connect your statistics to your research question.
Weak
"z = 3.45, p = .001"
Better
"Students in the intervention group showed significantly higher retention scores (M = 87.3, SD = 7.2) compared to the control group (M = 79.8, SD = 8.1), z = 3.45, p = .001, d = 0.98. This large effect suggests the intervention substantially improved learning outcomes."
See the difference? The second version interprets the findings and explains what they mean.
Non-Significant Results Still Matter
If your z-test wasn't significant, report it anyway. Null findings contribute to scientific knowledge.
"Contrary to our hypothesis, we found no significant difference in productivity between remote workers (M = 42.3, SD = 6.8, n = 180) and office workers (M = 41.7, SD = 7.1, n = 180), z = 0.73, p = .465."
