A semi-log graph is a type of graph that uses a logarithmic scale on the y-axis and a linear scale on the x-axis.
This type of graph is often used when the values for the y variable have much greater variability compared to the values for the x variable.
This occurs often in datasets in finance, economics, biology, and astronomy among other fields.
The following step-by-step example shows how to create a semi-log graph in Excel for a given dataset.
Step 1: Enter the Data
First, let’s enter the values for a fake dataset:
Step 2: Create a Scatterplot
Next, highlight the data values:
Along the top ribbon, click Insert. Then click the first option under the Scatter graph option:
The following scatterplot will automatically be displayed:
From the plot we can see that the values for the y variable have much higher variability than the values for the x variable.
This means it’s a good idea to convert the y-axis into a logarithmic scale to visualize the y-values more effectively.
Step 3: Modify the Y-Axis Scale
Next, right click the y-axis. In the dropdown menu that appears, click Format Axis:
In the window that appears on the right side of the screen, check the box next to Logarithmic scale:
The y-axis will automatically be converted to a logarithmic scale:
The x-axis remains on a linear scale, but the y-axis has been converted into a logarithmic scale.
Notice how much easier it is to interpret the y values in this graph compared to the previous graph.
Additional Resources
How to Create a Log-Log Plot in Excel
How to Transform Data in Excel (Log, Square Root, Cube Root)
How to Perform a Box-Cox Transformation in Excel