Because the range of total deaths varies significantly, it can be difficult to see the smaller counts on these graphs. Normalizing the data can improve visibility by removing drastic range differences. The two normalization options are log₁₀(x+1) and normalized. Log₁₀(x+1) transforms the data by adding one before taking the logarithm, which preserves zero values and mitigates right-skewed datasets, making them a more normal distribution. Normalized standardizes the data with extent normalization, which maps the minimum to zero and the maximum to one. Both make smaller values more visible, and make comparison easier.

Suggested citation

Please use the following as a suggested citation:

Savannah Scott and Jason Heppler, “Parish Deaths,” Death by Numbers, Roy Rosenzweig Center for History and New Media, George Mason University (2025): https://deathbynumbers.org/visualizations/sparklines/.