In this compelling episode of #ProveItPod, Dr. Todd Loushine and Dr. Matt Law deconstruct traditional safety measurement approaches. They challenge listeners to think critically about how safety is defined, measured, and understood, exploring the complex relationship between data, perception, and workplace safety.
Listen now at www.proveitpod.com or wherever you get your podcasts!
The presenters use real-world examples and academic research to illustrate how our current understanding of safety might be fundamentally flawed. They advocate for a more curious, experimental approach to safety management, emphasizing the importance of challenging assumptions and continuously learning.
Perfect for safety professionals, researchers, and anyone interested in understanding the deeper complexities of workplace safety beyond surface-level metrics.
Introduction of Speakers and Session Overview 00:00:48
The moderator, Kelton, introduced Dr. Todd Loushine and Dr. Matt Law as the presenters for the breakout session. Dr. Loushine was described as a certified safety professional with a PhD from Wisconsin, currently an associate professor at the University of Wisconsin-Whitewater, and a former compliance safety and health officer. Dr. Matt Law was introduced as a certified safety professional with a doctorate in public health from Walden University, whose dissertation focused on occupational risk perceptions among foreign-born construction workers in central Florida. Dr. Law mentioned his podcast "Prove It To Me," noting that the first full-length episode featured a conversation with Dr. Loushine on the same topics they would be discussing in this session.
Scientific Method and Research Fundamentals 00:04:34
Dr. Loushine began by explaining that the scientific method is cyclical rather than linear, often involving setbacks and sometimes yielding no results despite significant effort. He emphasized that research is about seeking truth and understanding, not just confirming existing beliefs. Dr. Loushine introduced key research concepts including construct validity (whether one variable truly relates to or causes another), measure reliability, and bias assessment. He stressed the importance of peer-reviewed publications versus non-reviewed content, noting a disconnect between researchers and practitioners in the safety field.
Importance of Curiosity in Safety Research 00:10:23
Dr. Matt Law emphasized the need for curiosity in safety research, encouraging the audience to question both ideas that align with and contradict their preconceptions. He noted that the safety and health field lacks consensus on the "right way" to do safety, with many approaches (BBS, HOP, Safety One, Safety Two, etc.) having little evidence-based support. Dr. Law framed this as an opportunity for testing and measurement to determine what actually works in improving workplace safety.
Challenging Traditional Safety Metrics 00:12:16
Dr. Loushine questioned how safety is defined and measured, suggesting that if something can't be measured, its effectiveness can't be established. He presented statistical concepts related to normal distribution and explained why injury data doesn't follow a normal distribution but is asymptotic. Dr. Loushine argued that comparing companies to industry averages using incident rates is problematic because extreme outliers skew the data. He questioned whether judging safety program performance based on injury numbers is appropriate, comparing it to judging driving performance based solely on accidents.
Research on OSHA Data Reporting Issues 00:27:16
Dr. Loushine shared findings from his published study analyzing seven years of OSHA 300A data, revealing that approximately 12% of submitted reports contained errors or inconsistencies in reported hours worked or employee numbers. These errors significantly impact incident rate calculations since hours worked is the denominator in the formula. After making corrections to the data, Dr. Loushine found that the calculated rates didn't match Bureau of Labor Statistics figures, raising questions about the validity of these metrics.
Reinterpreting Heinrich's Pyramid 00:32:41
The speakers discussed Heinrich's pyramid (the theory that for every major injury there are numerous minor incidents and near misses), arguing that Heinrich's work has been misinterpreted. Dr. Loushine explained that Heinrich never claimed these incidents were directly related but rather demonstrated the value in understanding near misses and minor incidents that occur more frequently. Dr. Matt Law noted that comparing different types of incidents using different calculation methods doesn't make sense but is often done because "it looks kinda sexy."
Correlation vs. Causation in Safety Research 00:35:15
Dr. Matt Law emphasized the distinction between correlation and causation, explaining that many factors contribute to safety outcomes, with any single factor typically accounting for only a small percentage of the variance. He used the example of organizational size correlating with technology adoption but only explaining about 20% of the variance. Dr. Loushine discussed the Hawthorne effect as an example of misattributing causation, where researchers initially thought lighting levels affected productivity but later discovered it was the attention workers received that made the difference.
Common Cognitive Biases in Safety 00:43:36
Dr. Loushine covered several cognitive biases affecting safety judgments: attribution error (blaming the person closest to an accident), equity theory (perceptions of fairness in the workplace), human error (the inevitability of mistakes), safety culture (organizational attitudes toward safety), and the Dunning-Kruger effect (overconfidence when lacking knowledge). He shared how understanding these biases helped him approach safety management differently in his part-time EHS manager role, creating more learning opportunities and better relationships with workers.
Interactive Exercise on Defining Safety 00:46:24
Dr. Matt Law led an interactive exercise asking attendees to define safety, receiving varied responses including "sending people home in the same condition," "presence of capacity for failure," "low recordable rate," and "consistently making the right choice." Dr. Loushine offered his definition: "Helping workers get their job done effectively with as minimal risk as possible and hopefully they elicit satisfaction from the work they do." Dr. Law highlighted that the lack of consensus on defining safety makes measuring it challenging.
Approaches to Safety Measurement 00:48:37
Dr. Law discussed the challenges of measuring safety when definitions vary widely. He shared an anecdote about being asked how many safety professionals an organization should have, noting that his research found no strong correlation between staffing levels and factors like employee count or revenue. He described an organization that successfully developed staffing models based on quantitative risk profiles of their sites. The speakers encouraged experimentation with different measurement approaches rather than relying solely on incident rates.
Conclusion and Final Thoughts 00:54:29
Dr. Loushine concluded with an Einstein quote about measuring a fish by its ability to climb a tree, emphasizing the importance of appropriate metrics. He encouraged attendees to "observe without bias, seek to understand and test your theories." Both speakers thanked the audience and invited them to connect on LinkedIn and listen to Dr. Law's podcast "Prove It To Me" for more research-based discussions on workplace health and safety.
