Use this plot to see if the effect of one variable on your outcome changes depending on the level of a second variable. It is a visual check to determine if variables act independently or if they work together to create a unique result in your data.
Visualizing Complex Relationships
A two-factor interaction plot helps you move beyond looking at variables one at a time. By fixing two variables and plotting their interaction against your outcome, you can quickly identify if they work together in a simple, additive way or if their relationship is more complex.
The plot uses one factor to define the X-axis and groups the second factor into individual lines. This structure allows for an immediate visual comparison across all combinations of these two variables.
Identifying Non-Additive Effects
In many datasets, the behavior of one variable is consistent regardless of external factors. However, when you observe lines that are not parallel, you have discovered an interaction. This suggests that the impact of your first factor is conditioned by the state of the second factor.
When lines actually cross, it points to a stronger effect where the trend for one group is effectively the opposite of another. Recognizing these patterns is essential for understanding how to adjust your inputs to achieve a desired output.
Data Integrity and Clarity
Lattice handles your data as-is, meaning we do not perform hidden interpolations or automatic filling of missing values. If a combination of factors lacks data, the tool explicitly marks these as empty cells.
This approach ensures that your visualization is a faithful representation of your underlying dataset. By maintaining transparency in how sparse data is handled, you can trust that the conclusions drawn from the plot are grounded in your actual experimental results.
1 · Intent → method
An LLM picks svt_plot_interaction from a fixed catalog.
2 · Method → numbers
Deterministic Python engine runs the math. Same input → same output.
3 · Numbers → plain language
A second LLM translates the result into your domain’s vocabulary.
How do I interpret the lines in this plot?
If the lines are parallel, the variables are independent and do not interact. If the lines are not parallel or appear to cross, it indicates that the variables have an interaction effect, meaning the influence of one depends on the level of the other.
What happens if there are missing combinations in my data?
This tool identifies empty combinations as 'sparse' cells. If your dataset lacks data for specific pairs of variable levels, the plot will reflect this without attempting to fill in those values, ensuring your analysis remains based on real observations.
Tool input schema
Schema for svt_plot_interaction not exported yet (run pnpm export:registry).