What is it about?

We chose three statistics that are commonly used as risk metrics (percentage, percentile, and ratio) and created a graph for each metric. We asked 676 people from the general public to read brief descriptions of a man convicted of a sexual offense, and gave them one of the metrics, with or without a graph. Participants who were given a risk metric and a graph perceived the man's risk to be higher when the risk was actually higher. The effect was larger when they had a graph and when percentiles were communicated.

Featured Image

Why is it important?

Most people are aware that courts make decisions about whether an individual is guilty or not. Yet most forensic decisions are about risk of future problems, not guilt. Risk is an important consideration for decisions about bail, probation, parole, levels of security and supervision, and so on. Understanding risk requires people to think in terms of probability (e.g., 30% likely) and comparison (e.g., twice as likely as average). This can be difficult, especially for people who don't have a lot of practice with mathematics and statistics. Our study shows that some graphs, especially with some statistics, can make it easier.


I hope this study helps make it easier for assessors to communicate risk more effectively and for decision makers to better understand and use risk assessment results. I enjoyed doing an experimental study for a change and I look forward to more results from Dr. Helmus's "Communication of Risk Assessment" (CORA) project.

Dr N Zoe Hilton
University of Toronto

Read the Original

This page is a summary of: Using Graphs in Sexual Violence Risk Communication: Benefits May Depend on the Risk Metric, Sexual Abuse, August 2020, SAGE Publications,
DOI: 10.1177/1079063220951191.
You can read the full text:




The following have contributed to this page