How do you choose the best type of plot to show interaction effects in your experiments?

1 Scatter plot

A scatter plot is a simple way to show the relationship between two continuous variables, such as height and weight. You can also use a scatter plot to show interaction effects by adding a third variable, such as gender, as a color or shape marker. For example, you can plot the height and weight of male and female students and see if there is a difference in the slope or pattern of the points. A scatter plot can help you identify outliers, clusters, and trends in your data, but it might not be very effective if you have too many points or too many levels of the third variable.

2 Line plot

A line plot is a good choice to show interaction effects when you have one continuous and one categorical variable, such as time and treatment. You can plot the mean or median of the continuous variable for each level of the categorical variable and connect them with a line. For example, you can plot the average blood pressure of patients who received a drug or a placebo over time and see if there is a difference in the shape or position of the lines. A line plot can help you compare the changes and differences between groups, but it might not be very accurate if you have a lot of variation or noise in your data.

3 Bar plot

A bar plot is a common way to show interaction effects when you have two categorical variables, such as gender and education. You can plot the frequency or proportion of one variable for each level of the other variable and stack or group them in bars. For example, you can plot the percentage of male and female students who have a college degree or not and see if there is a difference in the height or width of the bars. A bar plot can help you show the distribution and contrast between categories, but it might not be very informative if you have too many levels or too little variation in your data.

4 Facet plot

A facet plot is a useful way to show interaction effects when you have more than two variables, such as age, gender, and income. You can create multiple plots of the same type for each level of one variable and arrange them in rows or columns. For example, you can create scatter plots of income and education for male and female students and see if there is a difference in the correlation or dispersion of the points. A facet plot can help you explore the interaction effects across different subsets of your data, but it might not be very clear if you have too many facets or too complex plots.

5 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?