Why This Matters
When writing research papers or theses, it’s crucial to report statistical results clearly and consistently. This guide focuses on t-tests, covering descriptive statistics, p-values, and effect sizes.
The Standard Formula
The Anatomy of a Result
t (df) = test statistic, p = value, d = effect size
t-statistic
The calculated value from your t-test analysis.
Degrees of Freedom
Usually related to sample size (N - 1 or N - 2).
Significance
Probability that results occurred by chance.
Cohen's d
The magnitude (strength) of the effect.
Real-World Examples
"Students who used AI tools scored significantly higher (M = 78.2, SD = 4.5) than those who did not (M = 71.6, SD = 5.1), t(58) = 4.21, p < .001, d = 1.10."
- Report M and SD before the stats.
- Use italics for stats symbols.
"After using the chatbot, students’ post-test scores (M = 82.4, SD = 6.2) were significantly higher than pre-test scores (M = 75.0, SD = 5.9), t(29) = 5.13, p < .001, d = 0.94."
- Note the degrees of freedom changes.
- Round to 2 decimal places.
Table Format
Table 2
Independent Samples t-Test Results for Test Scores
| Group | M | SD | t(df) | p | d |
|---|---|---|---|---|---|
| AI Group | 78.2 | 4.5 | |||
| Control Group | 71.6 | 5.1 | 4.21 | <.001 | 1.10 |
Common Mistakes
Forgetting the effect size
Always report Cohen's d alongside p-values.
Too many decimal places
❌ p = 0.0000345
✅ p < .001
Reporting p = 0.000
A p-value is never exactly zero. Use p < .001.
Wrong emphasis
Do not use bold or underline for stats. Use italics for symbols (t, p, M, SD).
Summary Cheat Sheet
| Element | Format |
|---|---|
| Mean | M = [value] |
| Standard Deviation | SD = [value] |
| t-statistic | t(df) = [value] |
| p-value | p = [val] or < .001 |
| Effect Size | d = [value] |