A two-way ANOVA is used to determine whether or not there is a statistically significant difference between the means of three or more independent groups that have been split on two variables.
When reporting the results of a two-way ANOVA, we always use the following general structure:
- A brief description of the independent and dependent variables.
- Whether or not there was a significant interaction effect between the two independent variables.
- Whether or not the two independent variables had a statistically significant effect on the dependent variable.
Here’s the exact wording we can use:
A two-way ANOVA was performed to analyze the effect of [independent variable 1] and [independent variable 2] on [dependent variable].
A two-way ANOVA revealed that there [was or was not] a statistically significant interaction between the effects of [independent variable 1] and [independent variable 2] (F(df interaction, df within) = [F-value], p = [p-value]).
Simple main effects analysis showed that [independent variable 1] [did or did not] have a statistically significant effect on [dependent variable] (p = [p-value]).
Simple main effects analysis showed that [independent variable 2] [did or did not] have a statistically significant effect on [dependent variable] (p = [p-value]).
The following example shows how to report the results of a two-way ANOVA in practice.
Example: Reporting the Results of a Two-Way ANOVA
A botanist wants to know whether different levels of sunlight exposure and watering frequency effect plant growth. She plants 40 seeds and lets them grow for one month under different conditions for sunlight exposure and watering frequency.
She then performs a two-way ANOVA to determine if sunlight exposure and watering frequency effect plant growth.
The following table shows the results of the two-way ANOVA:

Here is how to report the results of the two-way ANOVA:
A two-way ANOVA was performed to analyze the effect of watering frequency and sunlight exposure on plant growth.
A two-way ANOVA revealed that there was not a statistically significant interaction between the effects of watering frequency and sunlight exposure (F(3, 32) = 1.242, p = .311).
Simple main effects analysis showed that watering frequency did not have a statistically significant effect on plant growth (p = .975).
Simple main effects analysis showed that sunlight exposure did have a statistically significant effect on plant growth (p < .000).
Things to Keep in Mind
Here are a few things to keep in mind when reporting the results of a two-way ANOVA:
1. Use a descriptive statistics table if necessary.
- It can be helpful to present a descriptive statistics table that shows the mean and standard deviation of values in each treatment group as well to give the reader a more complete picture of the data.
2. Round p-values when necessary.
- As a rule of thumb, the overall F-value and any p-values in ANOVA results are rounded to either two or three decimal places for brevity.
- No matter how many decimal places you choose to use, simply be consistent throughout the report.
Additional Resources
The following tutorials explain how to report other statistical tests and procedures in APA format:
How to Report One-Way ANOVA Results (With Examples)
How to Report Cronbach’s Alpha (With Examples)
How to Report t-Test Results (With Examples)
How to Report Pearson’s Correlation (With Examples)
How to Report Regression Results (With Examples)
This is a wonderful site! Thank you.
Hey Zach,
Thank you very much for this great and clear explanation! 🙂
May I ask: is this also transferable to reporting a three-way ANOVA? I’m trying to present the results of my 3-way ANOVA in a table, however, it’s quite overwhelming how to tackle this.
Kind regards,
Pia
Good day! May I know why the null hypothesis for Interaction was not rejected, since the F ratio (1.242) is greater than the p-value ( .311)?
Really helpful
Thank you Zach, it was really helpful
You are very welcome Sanghmitra!
I tested pH for five different samples for three weeks, should i perform two-way ANOVA or one-way ANOVA?
Hi Yesh…To determine whether to use a **two-way ANOVA** or a **one-way ANOVA**, let’s consider the structure of your data and the questions you want to answer:
1. **One-Way ANOVA**: Used when there is only one independent variable (factor) and you want to test its effect on the dependent variable (e.g., pH).
– Example: If you are only interested in comparing the pH of the five different samples (single factor: “sample type”), you would use a one-way ANOVA.
2. **Two-Way ANOVA**: Used when there are two independent variables (factors) and you want to test their individual and combined effects on the dependent variable.
– Example: If you want to examine how both **”sample type”** and **”week”** affect pH, and whether there is an interaction between these two factors, you should use a two-way ANOVA.
### Decision for Your Scenario:
– **Independent variables**:
– **Sample type (5 levels)**: Five different samples.
– **Week (3 levels)**: Three weeks.
– **Dependent variable**:
– pH measurements.
– **Questions to Answer**:
– Are there significant differences in pH between the different samples? (Main effect of “sample type”)
– Does pH change significantly over the weeks? (Main effect of “week”)
– Is there an interaction between sample type and week? (Interaction effect)
If you want to examine both the main effects (of sample type and week) and the interaction between them, you should use a **two-way ANOVA**.
### Assumptions of Two-Way ANOVA:
Before proceeding, ensure your data meets the following assumptions:
1. The dependent variable (pH) is measured on a continuous scale.
2. Observations are independent.
3. The groups are approximately normally distributed.
4. Homogeneity of variances (equal variances across groups).
If these assumptions are not met, you may need to use alternative methods (e.g., non-parametric tests).
Would you like guidance on performing and reporting the results of a two-way ANOVA?
how to report on interaction term in the two way or three way anaova
Hi Kowsi…When reporting the interaction term in a Two-Way or Three-Way ANOVA, follow these key points:
—
### **1. Two-Way ANOVA Interaction Reporting**
– **State whether the interaction effect is significant.**
– **Report F-value, degrees of freedom, and p-value.**
– **Describe the nature of the interaction using means, graphs, or post-hoc tests.**
#### **Example Format (APA Style)**
*”A two-way ANOVA was conducted to examine the effects of Gender (Male, Female) and Treatment (Control, Experimental) on Exam Scores. The interaction effect between Gender and Treatment was statistically significant, F(1, 56) = 5.72, p = 0.021, η² = 0.09. Simple effects analysis showed that males performed better in the experimental condition (M = 85.4, SD = 6.3) than in the control condition (M = 78.2, SD = 7.1), whereas females showed no significant difference between conditions (M = 82.3, SD = 5.9 vs. M = 81.7, SD = 6.5). These results suggest that the effect of treatment on exam scores depends on gender.”*
#### **Visualization**
– A **line graph** showing how one factor changes across levels of another.
– Report **simple effects analysis** to break down the interaction.
—
### **2. Three-Way ANOVA Interaction Reporting**
– **Mention any significant three-way interactions** (e.g., A × B × C).
– **Break down the interaction into two-way interactions and main effects.**
– **Include F-values, degrees of freedom, p-values, and effect sizes.**
– **Use plots and post-hoc tests to interpret results.**
#### **Example Format**
*”A 2 × 2 × 2 ANOVA was conducted to examine the effects of Age (Young, Old), Gender (Male, Female), and Exercise Type (Aerobic, Strength) on Endurance Scores. The three-way interaction among Age, Gender, and Exercise Type was significant, F(1, 120) = 7.45, p = 0.008, η² = 0.12, suggesting that the interaction between Gender and Exercise Type differed across Age groups. Post-hoc analysis revealed that for younger participants, males benefitted more from aerobic exercise (M = 78.5, SD = 5.3) than strength training (M = 71.2, SD = 6.1), whereas no such difference was found for females (p = 0.68). In older participants, both genders performed better with strength training compared to aerobic exercise. These results indicate that the effectiveness of exercise type is influenced by both age and gender.”*
#### **Visualization**
– A **3D interaction plot** (if applicable).
– Multiple **line graphs** showing interaction effects within subgroups.
—
### **General Reporting Guidelines**
– **Use APA format:** Include F(df, df), p-values, and effect sizes (η² or partial η²).
– **Provide context:** Explain what the interaction means in practical terms.
– **Use post-hoc tests:** Simple effects analysis or pairwise comparisons help clarify the interaction.
– **Include visualizations:** Graphs make interactions easier to interpret.
Hi
I conducted a 2-armed (control and intervention) randomized trial to test the effect of a drug on quality of life (QoL). I measured QoL 5 times during one year. I used GraphPad 2-anova with repeated measures, and got this result-table:
Source of Variation % of total variation P value P value summary Significant? Geisser-Greenhouse’s epsilon
Time x Column Factor 2,444 <0,0001 **** Yes
Time 4,432 <0,0001 **** Yes 0,8366
Column Factor 8,553 <0,0001 **** Yes
Subject 47,56 <0,0001 **** Yes
ANOVA table SS DF MS F (DFn, DFd) P value
Time x Column Factor 18247 4 4562 F (4, 448) = 7,396 P<0,0001
Time 33092 4 8273 F (3,346, 374,8) = 13,41 P<0,0001
Column Factor 63861 1 63861 F (1, 112) = 20,14 P<0,0001
Subject 355154 112 3171 F (112, 448) = 5,141 P<0,0001
Residual 276324 448 616,8
Difference between column means
Mean of Group A 51,78
Mean of Group B 30,61
Difference between means 21,17
SE of difference 4,717
95% CI of difference 11,82 to 30,52
How would you report this in a journal?
You’re doing great — you’ve clearly structured the randomized controlled trial, analyzed it properly with **2-way repeated measures ANOVA**, and you have rich results.
Here’s **how you could report your findings formally in a journal**, step by step, following typical biomedical/scientific writing standards:
—
### Statistical Analysis and Results
A two-way repeated measures ANOVA was conducted to evaluate the effects of the intervention (drug) versus control over time on quality of life (QoL) scores, measured at five time points across one year.
There were statistically significant main effects of:
– **Time** (F(3.35, 374.8) = 13.41, *P* < 0.0001), - **Group** (F(1, 112) = 20.14, *P* < 0.0001), - and a significant **Time × Group interaction** (F(4, 448) = 7.396, *P* < 0.0001). The **Geisser-Greenhouse correction** was applied to adjust for potential violations of sphericity (ε = 0.8366). Analysis of the differences between groups showed that the mean QoL score was significantly higher in the intervention group (mean = 51.78) compared to the control group (mean = 30.61), with a mean difference of 21.17 points (95% CI: 11.82 to 30.52, SE = 4.717). The subjects' variance accounted for 47.56% of the total variability (*P* < 0.0001). These results suggest that the intervention significantly improved QoL over the study period compared to the control group, with changes in QoL also varying over time between the groups. --- ### Shorter Summary Version (e.g., for Abstract) > A two-way repeated measures ANOVA revealed significant main effects of group (*P* < 0.0001) and time (*P* < 0.0001), as well as a significant time-by-group interaction (*P* < 0.0001), indicating that changes in QoL over time differed between groups. The intervention group demonstrated higher QoL scores than the control group (mean difference 21.17 points, 95% CI 11.82 to 30.52). --- ### Notes for Reporting: - Always **report F-statistics with degrees of freedom**, the **F value**, and the **P value**. - Since you applied the **Geisser-Greenhouse epsilon correction**, mention that, because it shows you checked for violation of sphericity. - **Confidence intervals (CI)** for the mean difference give more information than only p-values — you correctly included them. - The **percentage of variance** (like "Subject 47.56%") is usually secondary, but you can mention it if the journal style permits. ---