Understanding Interaction Plots: A Step-by-Step Guide Using Excel

The interaction plot is a powerful graphical tool used in statistical analysis to visualize how two or more independent variables influence a single dependent variable. This visualization is particularly useful in experimental design, where researchers seek to understand complex relationships beyond simple averages.

A well-constructed interaction plot displays the mean value of the outcome measure (the dependent variable) across all combinations of the levels associated with the independent factors. By examining the pattern and slope of the lines on the plot, we can quickly determine if an interaction effect is present—meaning the effect of one factor changes depending on the level of the other factor.

This detailed, step-by-step tutorial demonstrates exactly how to create and interpret a high-quality interaction plot using Microsoft Excel, a widely accessible tool for data visualization. We will walk through the entire process, from setting up your experimental data table to drawing conclusions about the effects of your variables.

Defining Key Statistical Concepts

Before diving into the Excel mechanics, it is essential to clarify the three main types of effects we look for when analyzing data using an interaction plot: the main effects and the interaction effect. Understanding these concepts ensures accurate interpretation of the resulting chart.

A main effect refers to the overall effect of a single independent variable on the dependent variable, averaging across the levels of the other independent variable(s). If one factor consistently shows a strong influence regardless of the second factor, it demonstrates a strong main effect.

Conversely, the interaction effect occurs when the impact of one factor depends on the level of the second factor. Graphically, this is represented by non-parallel lines on the interaction plot. If the lines cross or diverge significantly, an interaction is likely present, indicating a synergistic or antagonistic relationship between the factors.

Step 1: Setting Up the Experimental Data

For this example, we will investigate the factors that contribute to plant growth. Specifically, we are interested in how two independent variables—sunlight exposure (at two levels: low vs. high) and watering frequency (at two levels: daily vs. weekly)—affect the overall plant growth (our dependent variable).

Imagine we conducted an experiment using 60 plants, dividing them equally among the four treatment groups. We calculated the mean plant growth (in inches) for each combination of sunlight and watering frequency. These crucial mean values must be organized logically within Excel to facilitate chart creation.

The image below illustrates the correct layout for entering these summary statistics into your worksheet. This specific format ensures that Excel correctly maps the factor levels to the line chart axes and data series.

From this data table, we can immediately observe the preliminary mean outcomes for each treatment combination:

• Plants receiving high sunlight and daily watering averaged 8.2 inches of growth.
• Plants receiving high sunlight and weekly watering averaged 9.6 inches of growth.
• Plants receiving low sunlight and daily watering averaged 5.3 inches of growth.
• Plants receiving low sunlight and weekly watering averaged 5.8 inches of growth.

Step 2: Generating the Interaction Plot in Excel

Creating the graph requires careful selection of the data range. We must highlight only the cell range containing the factor levels and the calculated mean values—in this specific setup, that corresponds to range C4:E6. Excluding the overall group labels prevents Excel from mistakenly plotting those cells as a separate data series.

Once the appropriate range is highlighted, navigate to the Insert tab located on the top ribbon of Excel. Within the Chart group, select the Line Chart option. We recommend choosing the first standard 2-D Line chart type, which plots the mean values as a continuous line across categories.

Excel will instantly generate the initial line chart, which serves as our preliminary interaction plot. You may need to clean up the chart title and axis labels for presentation purposes, but the core visualization is immediately ready for interpretation.

This plot visually represents the means, allowing us to proceed to the crucial stage of analysis, where we assess the relationship between the factors.

Step 3: Interpreting the Main Effects

The primary goal of creating an interaction plot is to visually assess the existence and magnitude of the main effects and the interaction effect. We begin by interpreting the vertical distance between the lines to understand the overall influence of each factor.

Main Effect of Watering Frequency: To assess the main effect of watering frequency, observe the overall flatness or slope of the two lines relative to the X-axis points (Daily vs. Weekly). Since the lines appear relatively flat across the X-axis, the difference in plant growth between daily and weekly watering is minimal. This visual evidence suggests that watering frequency alone likely does not have a statistically significant effect on plant growth.

Main Effect of Sunlight Exposure: To assess the main effect of sunlight, observe the vertical distance between the two lines (representing Low Sunlight and High Sunlight). The lines are distinctly separated, indicating a substantial difference in growth means. Plants under high sunlight (average growth 8–9 inches) grew considerably more than those under low sunlight (average growth 5–6 inches). This large vertical separation strongly suggests that sunlight exposure does have a statistically significant main effect on plant growth.

Analyzing the Interaction Effect

The most critical piece of information conveyed by this plot is the presence or absence of an interaction. An interaction effect is indicated when the lines are non-parallel—meaning the slope of one line is different from the slope of the other.

In our plant growth example, the two lines representing the high and low sunlight conditions are visibly parallel. They maintain a consistent vertical distance between them as they move from “Daily” to “Weekly” watering. This parallelism is the key visual cue that confirms there is no interaction effect between watering frequency and sunlight exposure.

This finding means that the influence of sunlight on plant growth is constant, regardless of whether the plant is watered daily or weekly. Similarly, the marginal effect of watering frequency remains the same whether the plant is exposed to high or low light. Had the lines crossed or significantly converged, we would conclude that the effect of one variable is dependent on the level of the other.

Conclusion and Further Reading

The ability to create and interpret an interaction plot in Excel is a fundamental skill for anyone analyzing experimental data. While the graph provides a powerful visual summary, remember that formal statistical testing (such as ANOVA) is required to definitively confirm whether the effects observed are truly statistically significant.

In summary, our plot confirmed a strong main effect of sunlight exposure, a weak main effect of watering frequency, and, critically, the absence of an interaction effect between the two factors on plant growth.