In safety management, data is often treated as objective truth. Leaders use incident rates, near-miss reports, injury trends and predictive models to guide them as they prioritize risk and allocate organizational resources. Yet data can quietly mislead us, particularly when bias is embedded in what we collect and our measurement and interpretation methods. Effective, ethical safety leaders continuously work to recognize and address these distortions. Exploring Various Biases A widely cited World War II-era example illustrates the dangers of biased data. During the war, Allied forces studied returning aircraft to determine where additional armor was needed. Analysts initially recommended reinforcing areas with the most visible bullet holes. Statistician Abraham Wald challenged this reasoning, highlighting what is now known as survivorship bias. He observed that the only aircraft analyzed were those that survived their missions. Aircraft that failed to return home had likely sustained catastrophic damage to areas where no bullet holes were observed on the surviving planes. Wald’s insight suggested that undamaged areas required reinforcement, not the visibly damaged ones. Survivorship bias remains a powerful warning for leaders whose safety programs rely on incomplete or filtered data. However, it is only one source of potential distortion. Selection bias occurs when data is drawn from an unrepresentative sample. In a utility environment, this could happen when organizations heavily rely on information from crews or regions with strong reporting cultures while underestimating risk in areas where incidents and near misses are less likely to be reported. Leaders may inadvertently prioritize the wrong hazards when the dataset does not reflect the entire population. Even when data is broadly collected, confirmation bias can still emerge (i.e., leaders subconsciously favor data that supports their existing beliefs or assumptions). For example, if management believes a particular work practice is safe, near-miss data that challenges their belief may be discounted or dismissed as anomalous. Over time, selective interpretation reinforces blind spots and weakens organizational learning. Measurement bias can be introduced at the point of data capture, resulting in inconsistently defined or poorly standardized safety data. Metrics that depend on subjective judgment – such as what qualifies as a safety observation or near miss – can vary widely among supervisors, crews and contractors. When measurement practices differ, trends become unreliable and comparisons across departments or time periods lose meaning. Historical bias arises when data reflects outdated assumptions, norms or exclusions that no longer align with today’s workforce or operating environments. Caroline Criado Perez’s book “Invisible Women: Data Bias in a World Designed for Men” highlights how systems built on incomplete data can overlook entire populations. In safety-critical industries, this could appear in PPE design, equipment ergonomics or training materials developed for a narrow segment of the workforce, leaving others at elevated risk. More recently, algorithmic bias has emerged as organizations increasingly adopt predictive analytics and other safety tools driven by artificial intelligence, which can inherit and amplify patterns embedded in historical data. Any algorithms trained using past incident data that underrepresents certain hazards, job roles or worker groups may consistently underestimate risk in those areas. Since algorithmic outputs often appear objective, this bias can be difficult to detect and challenge without deliberate oversight. Overcoming Vulnerabilities Embedded bias distorts safety intelligence and can create organizational vulnerabilities. Resources may be misdirected. Early warning signs could be missed. Emerging hazards might remain invisible until a serious incident occurs. Overreliance on lagging indicators like recordable injury rates could create a false sense of security, especially in high-risk utility operations. Biased data can also further erode trust. Reporting declines when frontline workers witness leadership decisions that conflict with their lived experiences, deepening the data gap. Despite these risks, high-quality data remains indispensable to effective safety management, enabling organizations to identify trends, prioritize controls, evaluate interventions, and shift from reactive responses to proactive prevention. Decisions made without data are often driven by anecdotes and intuition. The challenge, therefore, is not whether to use data but how to use it thoughtfully and with full awareness of its limitations. Recognizing bias is the first step. Leaders should routinely ask, who is missing from this dataset? What assumptions shaped these metrics? What risks could be hidden? A questioning approach encourages more accurate, proactive, ethical decision-making. Leaders who understand bias are more likely to consult multiple data sources, blending quantitative indicators with qualitative insights from job observations, worker feedback and learning teams. Most importantly, confronting data bias helps to ensure that safety systems are designed to protect all workers, not just those most visible in the data. Intentional effort is required to turn awareness into action. Organizations must routinely audit safety data for gaps and inconsistencies, standardize definitions and measurement practices, and foster psychologically safe reporting environments. As use of predictive analytics and other AI tools expands, transparency and human oversight are essential. Leaders must treat model outputs as decision aids – not decision-makers – and be accountable for how data-driven insights are applied in the field. Conclusion Numbers carry authority, shaping organizational budgets, priorities and narratives. However, as Abraham Wald demonstrated decades ago, some of our greatest threats may never appear in the data we see. Safety leaders who understand and deliberately question, test and correct for biases ultimately position their organizations to more effectively mitigate risk. About the Author: Gina Vanderlin, CSP, CHMM, CIT, CUSP, is the customer operations health and safety program manager at PSEG Long Island. With over 15 years of experience leading EHS initiatives in high-reliability industries, she remains passionate about elevating safety from a compliance function to a strategic driver of culture, engagement and operational excellence. Reach Vanderlin at gina.vanderlin@psegliny.com.

I’m not sure how I became an analyst. It wasn’t something I planned for. Various types of analyst roles exist, but I primarily analyze incidents, breaking down and studying the elements of events to identify causes and effects. Incident analysis, done well, ultimately helps prevent undesired future outcomes. Over the last 15 years, I have analyzed a half-dozen apprentice training yard accidents and watched two videos of apprentice-involved incidents. These events are reminders that lineworkers frequently learn their lessons the hard way. I continue striving to change that fact because – far too often – the hard way becomes the final act to what could have been a great life. I was once asked to write an opinion about a root cause analysis (RCA) conducted by OSHA and a utility. The analysis focused on a singular event that put three apprentices in the hospital. OSHA performs RCAs only to identify whether employers are at fault. The analysis I was asked to write about stated that the incident’s cause was various physical conditions and procedural mistakes. But while the conditions and mistakes were causally related, none was the true root cause. That concerning realization is the reason I wrote this article: to clarify what a good RCA entails and explore its relationship with lessons learned from training accidents. A Peculiar Art Form RCA is a peculiar art form that requires analysts to be knowledgeable about safety standards and human performance principles. Numerous utilities use RCA software applications, mostly algorithm-based methodologies designed to help investigators determine the most likely root cause. The applications were developed to standardize RCAs, offering guided protocols to prevent investigator errors. However, the human element can still impact results. A persistent issue with RCA application use is listing, evaluating and interpreting the causal factors that preceded an incident. Causal factors contribute to the incident under investigation, but they are not the root cause. The root cause is the singular event that prompted the incident; if it had not occurred, the incident would not have occurred either. I recently reviewed two incidents so similar in nature that the same investigation report could have been written for both. In these cases, which took place a few years apart, investigators used RCA software to determine a root cause. The only difference between the two final RCA reports? You guessed it: the identified root cause. Two entirely different RCA conclusions resulted from the very same causal factors. My point here is that an RCA application is only effective when users complete the software training and stick to its process. Root causes are not always easy to determine, and they are not always what we initially believe they are – which brings us back to training. Introducing Thom and Goob To demonstrate rodeo-style hurt man rescue, an apprentice named Thom climbed to the top of a distribution pole. He successfully reached the mannequin only because of his portable fall protection device. Thom then fumbled with the rigging, desperately trying to get the mannequin down in four minutes. He could hear encouragement from the ground, shouts of “Go, go, go,” “Wrap this,” “Pull that,” “Reach around that.” Finally rigged, Thom reached around the mannequin and, using the hawkbill knife that he had sharpened to a razor’s edge for the demonstration, cut his own fall protection. He fell 38 feet, right into life in a wheelchair. In another rescue incident (see https://youtu.be/gaH7pK-6n84), a worker nicknamed Goob also inadvertently cut his fall protection. I don’t know how that worked out for him, but for lineworkers reading this, the lesson is found in Goob’s now-infamous rescue fail. We have an industry training shortcoming that is exacerbated by our need to get lineworkers trained and in the air (or in the ditch). The problem lies in the difference between objective and subjective training goals. Earlier, I stated that Thom’s portable fall protection device was the sole reason he reached the mannequin. Neither Thom nor Goob was subjectively competent enough to successfully complete their tasks. The difference between subjectivity and objectivity plays a considerable role in training, particularly when training trainers. Subjectivity vs. Objectivity Objective evaluation is rooted only in facts and goals. Subjective evaluation is influenced by the evaluator’s personal experience, feelings and opinions. In this context, “personal experience” is legitimate hands-on utility industry experience. Note: To be clear, I believe that good instructors possess a great amount of career experience and industry knowledge. This is not an indictment of on-the-job (OTJ) training using lineworker mentors. I am a product of the OTJ process and have great respect for those who taught me. Negligence and incompetence are not the issues I am addressing here. The problem is generic in nature and perhaps even a hidden organizational defect. Thom, the apprentice who made it to the mannequin, fell victim to two objective influences that resulted in his fall. First, he was not a competent climber. It is unlikely that Thom would have climbed a 40-foot pole had he not been wearing a personal fall arrest system (PFAS). He was allowed to do so because he had trained in a PFAS that everyone believed would prevent his fall. Using the system, Thom got to the top of the pole, but his hook sets were tentative. His body was off-center and uncomfortably oriented because he did not periodically adjust his PFAS during ascent. The “git-r-done” mentality was the second objective influence. Although I love 1990s-era Larry the Cable Guy, he didn’t do us any favors, but it’s not really his fault. Git-r-done was a comedy phenomenon that made light of simple men using unsophisticated methods to complete manly tasks, resulting in their unrestrained celebration. I am in favor of all those things, especially the unrestrained celebration, but the industry may have taken git-r-done too far. Encouragement from people on the ground was an additional event precursor in both Thom’s case and Goob’s case. Objectively, both apprentices were working hard to succeed; it is human nature to seek approval from others. But the level of problem-solving Thom and Goob displayed demonstrated that the two men did not possess the competence needed. Goob’s pole strap was too far out, and he displayed poor foot positioning and poor rigging management skills. Thom’s circumstances were the same. He explained to me that he had not felt confident in his climbing skills while on the pole but believed his PFAS would protect him. When Thom reached the top, his problem-solving skills were compromised by his lack of experience and the pressure of well-intentioned coworkers shouting encouragement from below. Competence is the first goal of industry training. Next, trainees are coached to both competently and efficiently complete their tasks. Until an apprentice demonstrates adequate problem-solving and skill competency independent of trainer instruction, the process must be unrushed and orderly. An apprentice simply climbing to the top of a pole is an objective measurement of quality. Climbing to the top with demonstrated skill is a subjective measurement. Where Do RCAs Fit In? Trainer competency was the root cause of Thom’s incident, Goob’s incident and the other training yard incidents I referenced earlier. But that is not because the trainers were incompetent. Rather, they had not been sufficiently trained to train other workers. I hate making that statement without first preparing readers because these incidents truly were not the fault of industry trainers. They did not lack lineworker skills or knowledge; they lacked understanding of the individuals who they were training. An effective trainer understands the nature of the trainee and recognizes subjective indicators of their competence to safely perform learned skills. Those trainers with effective technology transfer skills understand the nature of the learning and the learner, training modalities, subjective indicators of training success and objective competency measures. Falls accounted for two of the previously referenced training yard incidents. In both instances, the instructors bowed to trainee pressure, deviating from their planned training methodologies to instead oversee speed-climbing events. One apprentice climbed off the top of a pole. The other gaffed himself and almost bled to death. Neither had climbing skill characterized by good hook sets, technique, hand position and body orientation. In fact, neither even looked up while ascending. In yet another case, apprentices used trial and error to learn how to compress sleeves with a 60-ton press. Blown hydraulic hoses hospitalized two individuals. The utility’s investigation blamed a failed hose that “should have been capable of containing the hydraulic pressure.” But the RCA revealed that the hose had been plugged into a universal pump open-center system, which prematurely forced pressure into the hose of the closed center head, preventing operation of the spring-loaded quick coupling. The apprentices forced the coupling and broke the hose. So, the problem wasn’t a faulty hose but an objective-based training issue: “Squeeze the sleeve without bending it.” That initial introduction to the task and the training methodology lacked requisite preparation and instruction elements, which should have been audited by observing the steps via an orderly process. The Bottom Line I am now going to use RCA to lobby for a comprehensive review of the nature of our industry’s training. We have too many examples of trainees pushed beyond their skill level, and they are not limited to climbing in 100% fall protection. Let’s audit our training processes, including how we train our trainers, ensuring that we provide them with the technology transfer skills they need to successfully pass on their experience, craft skills and other knowledge. A utility manager once challenged me about the cost of training the organization’s trainers. He said it was frustrating to spend the time and money just for them to eventually leave. His attitude is one more example of mistakenly relying on objective concerns. His subjective concern should have been, what if we don’t train our trainers and they stay? About the Author: After 25 years as a transmission-distribution lineman and foreman, Jim Vaughn, CUSP, has devoted the last 28 years to safety and training. A noted author, trainer and lecturer, he is a senior consultant for the Institute for Safety in Powerline Construction. He can be reached at jim@ispconline.com.