By Matthew S. O’Connell and Kristin Delgado
Use of targeted selection systems can enhance the effectiveness of organizational safety initiatives by selecting, at the individual level, employees who are more likely to behave in a safe manner. Such targeting also helps select leaders who are better suited to promote a safe organizational climate.
Despite continuing efforts to improve safety climate and greater sophistication in ergonomics engineering, the fact is that approximately 40 percent of work-related accidents can be attributed to a failure to use the protective gear provided at the workplace. This statistic has not decreased significantly during the past 20 years. What are the reasons for the failure to use appropriate protection devices and follow safety guidelines? And, more importantly, what can organizations do to help promote employee adherence to safety rules and procedures? A dual-focused strategy of employee selection practices, both at the employee level and at the leadership level, can help reduce workplace accidents and enhance the overall organizational safety climate.
Organizational culture plays an important role in the encouragement and reinforcement of safety guideline compliance among employees. Some organizations hinder workplace safety efforts by placing a higher emphasis on productivity than on safety. For example, the root cause of recent catastrophic mining accidents has been traced back to a blatant disregard for the safety of the employees on the part of management in the hopes of extracting every ounce of productivity out of the mine.
However, bear in mind the individual worker plays a key role in workplace safety. Carelessness, recklessness and naiveté on the worker’s part all contribute to poor safety compliance and, ultimately, injuries. Some of this can be corrected through training. A lack of knowledge about safety practices is related strongly to safety violations and accidents. This suggests that some accidents occur because the worker was not prepared sufficiently, was not aware of the proper equipment to use, or was not aware of the correct procedures to follow.
Employee selection can play a crucial role in screening out individuals who are more likely to pose a risk to workplace safety. Research shows that some individuals are more likely to engage in high risk, unsafe behaviors than others. For instance, 20 percent of drivers account for almost 80 percent of all driving accidents. That 80/20 pattern is similar to what is found in other industries for negative behaviors, including accidents, injuries and safety violations. Thus, the assumption that most accidents are simply the product of the environment, poor training or poor safety climate ignores the individual’s role in accident prevention.
Individual characteristics related to safety performance
In order to identify and select the individuals who are more likely to be safe workers, it is important to distinguish why some people are more likely to be involved in safety incidents. Research points to key individual-level factors. Safety performance can be defined as safety-related behaviors, such as following safety procedures, using safety equipment and encouraging others to follow rules, whereas safety outcomes are tangible events, such as accidents or injuries.
Across occupations, acting responsibly is of utmost importance. Following the rules, remaining on-task, working hard and making well-thought-out judgments all contribute to being a safe worker. A key factor that contributes to these behaviors is the trait of conscientiousness. People who are highly conscientious are hard workers; they want to do the right thing and are more apt to follow rules and follow through. As such, they are much more likely to be safe employees. Individuals who are low in conscientiousness may disregard rules and, in some cases, actively rebel against authority. Research has shown that these behaviors are associated with higher traffic violations and unsafe work behaviors.
Locus of control is another personality trait that can differentiate people who act responsibly from those who do not. Individuals with an internal locus of control believe that they have control over what happens to them. As such, they are much more likely to take action to prevent negative events such as accidents or equipment failure. In contrast, individuals with an external locus of control perceive that many things in life are out of their hands, including safety issues. Therefore, these individuals may not intervene or take action when needed. Across a variety of occupations, locus of control has been found to predict accident risk, number of reported accidents and accident severity. Moreover, according to some research, employees with an external safety locus of control reported significantly more accidents, which led to significantly higher medical costs.
A third factor, impulsivity, which is closely related to thrill seeking and risk taking, long has been associated with unsafe behaviors. Impulsive individuals tend to be more volatile, unpredictable and less adaptive to change. They have difficulty managing stress and are more likely to take unreasonable chances, ignore safety rules and make rash decisions that may place themselves and others in compromising and potentially dangerous situations. These individuals are more likely to engage in unsafe behaviors because they underestimate the likelihood of getting hurt, or they enjoy the thrill of taking risks.
The combination of the key factors of conscientiousness, locus of control and impulsivity can help determine an accurate and consistent method of screening individuals for jobs that have a higher possibility of accidents or physical injury. The following studies illustrate the effectiveness of using such an approach in the hiring process.
Study 1: Reducing injuries in the auto industry
Injury data from a Canadian-based auto manufacturer over a one-year period was tracked for employees hired using a new selection process, which included the combination of conscientiousness, locus of control and impulsivity. Results were compared to injury data from employees hired prior to use of this new selection process. During this one-year period, 497 individuals from a group of 3,932 employees (12.6 percent) hired prior to the new selection process reported an injury of one type or another. During that same time period, 10 out of a group of 294 individuals (3.4 percent) hired using the new process reported injuries. See Figure 1.
A limitation of this study was that the injury data did not differentiate between acute and chronic injuries. People employed longer likely will exhibit more chronic, repetitive motion injuries than newer employees. However, the organization reported that injuries within the first year of employment, especially acute injuries, are more typical than injuries among more experienced employees. The limitation of the study notwithstanding, it is clear that there was a significant reduction in reported injuries among those hired using the new screening process.
Study 2: Identifying accident risks early in the hiring process
This study asked 730 candidates applying for hourly production jobs at a large U.S. manufacturing firm to report how many work-related safety incidents they had been involved in within the past five years. Recommendations were made based on the results of the assessment scores. A weighted composite – primarily composed of the three factors of conscientiousness, locus of control and impulsivity – resulted in four possible categories. On the low end, the grouping of “not recommended” consisted of those with a high risk for engaging in unsafe behaviors, or people less likely to be productive employees. The “strongly recommended” tier included those with a low risk for engaging in unsafe behaviors or people more likely to be productive employees.
As can be seen in Figure 2, 22 percent of the candidates who were “not recommended” reported that they had been involved in at least one safety incident in the past five years. On the other hand, only 7 percent of candidates in the “strongly recommended” responded so.
In addition, we asked candidates how likely they would be to report co-workers who break safety rules. This specific safety performance behavior of promoting safety initiatives has been identified as an important aspect of safety compliance. Employees willing to report others who violate safety rules are more likely to participate in safety behaviors and promote a better overall safety climate at work. As Figure 3 shows, only 23 percent of “not recommended” candidates reported that they would report rule violations. This is quite low compared to the likelihood of reporting safety violations of 49 percent in the “recommended” category and 73 percent in the “strongly recommended” category.
There are limitations with this methodology. First, it relies on self-reports of past behaviors regarding previous accidents and behavioral intentions of reporting someone who breaks the rules. However, several studies have shown that self-reported accidents and supervisor ratings of accidents correlate quite strongly. Moreover, there is a large body of research on attitudes, intentions and behaviors that demonstrates that intentions are significant and strong predictors of future behaviors. The presumption here is not that involvement in one work-related accident will result in another future accident. To be sure, myriad factors lead to an accident, many of them out of the control of the individual. Nonetheless, based on a study of more than 500,000 commercial drivers, involvement in a prior crash increased the likelihood of a future crash by 87 percent.
Second, this study did not control for differential rates of exposure to dangerous situations. This is difficult to control in an applicant pool where candidates come from differing backgrounds and have different levels of experience. Nonetheless, nothing leads us to think that there would be a significant relationship between previous levels of exposure to unsafe environments and the three primary factors of conscientiousness, locus of control and impulsivity that drove the recommendations on the predictor side.
Taken together, the results of these studies, in conjunction with a long list of other applied studies, demonstrate that screening candidates on characteristics associated with increased risk for unsafe behavior is an effective and straightforward strategy to reduce risk in the workplace. Interestingly, the characteristics of those less likely to be involved in accidents are the same factors associated with higher levels of productivity. So focusing on these characteristics and other important job-related factors while hiring not only reduces the likelihood of accidents and workers’ compensation claims, it might improve productivity, quality and organizational citizenship behavior.
Hiring the right leaders
Clearly, hiring the right employees and placing them in a poor environment under a bad leader is a recipe for disaster. The best employees deserve and excel under solid leadership, and, in fact, they expect it. This is just as much the case when the focus is on improving work force safety and reducing accidents as it is for improving productivity and quality. A substantial body of research history demonstrates the effectiveness of interventions designed at improving safety climate and behavioral safety. These interventions tend to yield quite positive results. In other words, training leaders and employees to focus on safety behaviors, placing a stronger focus on safety performance, and providing tools and feedback to help the process along are effective means of reducing accidents.
The relationship of leadership to safety performance, safety commitment and safety outcomes has been studied extensively over the past 20 years, resulting in consistent and important findings. The vast majority of research has focused on subordinates’ perceptions of leaders and the perceived commitment to safety on the part of management as the predictor of safety outcomes. The research shows that the leaders’ communication style, the way they influence, the level of empowerment or decision making they allow subordinates, and their commitment to subordinates’ health and welfare appear to be related to safety performance, perceived safety climate, commitment to follow accepted safety procedures and regulations, and, to a lesser extent, accidents.
While there is a growing level of agreement regarding which styles of leadership foster better safety performance, there is a dearth of research on how best to select these leaders. Cognitive ability typically is seen as the single best predictor of performance, especially in higher level and more complex jobs. This makes sense. Smarter people process information more quickly and are more likely to make better decisions. However, leader cognitive ability rarely is mentioned as a contributing factor in the research on safety outcomes and safety leadership. This may have something to do with the primary research methodology employed in such research, where the focus is primarily on subordinates’ perceptions of their leaders. Thus, while it is clear that perceptions of leadership style are related to safety outcomes, there is not a direct link between leadership style and cognitive ability. Cognitive ability is likely a necessary but not sufficient factor in hiring the right leader.
To that end, we have developed a robust assessment solution for screening front-line leaders into positions where safety is a primary concern. This went far beyond simply measuring cognitive ability and instead focused on 11 competency areas previously shown to be related to job performance, along with a new competency, which we referred to as “safety leadership.” The assessment was constructed to include a number of different types of assessment methodologies, including realistic situational judgment scenarios, personality, applied logical reasoning, problem-solving simulations and a number of interactive simulations that measured multitasking.
Leadership study 1: Global mining company
The opportunity for catastrophic accidents is particularly high in the mining industry. On a daily basis, individuals are faced with the possibility of serious physical injuries. We administered the leadership assessment to 85 current managers in a large multinational mining company from 11 separate mining locations in seven different countries. Performance ratings for these individuals were completed by their direct supervisors. Unfortunately, in this particular study no information on safety or accidents was available. Nonetheless, we were interested in seeing how well a leadership-level assessment would predict job performance in an environment where safety was at a premium.
The results were quite positive. Both the uncorrected and corrected correlation between the overall assessment score (average of all competencies) and the overall performance rating was strong, significant and positive. This demonstrated how an objective assessment could predict job performance accurately across countries and locations for leaders in a demanding and dangerous work environment.
Leadership study 2: Global explosive manufacturer
If there is an industry that is potentially more dangerous in terms of serious injury, it probably is the manufacture and deployment of explosives, which are used extensively in the mining industry. In this study, a group of 39 individuals, 27 team leaders and 12 managers from the company’s Swedish location were assessed using the leadership assessment tool described above. Based on an analysis of the jobs in question, a “footprint” was established for the assessment. The footprint essentially established acceptable levels of performance on each of the 11 competencies, as well as a lower limit for each competency. Actual safety data was gathered from 2000 to 2008 for the managerial group. That data included four separate types of incidents ranging from less severe to more severe. The four safety incident types were general learning incidents, minor incidents, recordable incidents, and fire or explosion incidents. These criteria also were combined to create a total number of incidents. The data represent average number of incidents per year for the group of individuals supervised by these managers. Using a composite (average) of the 11 competencies as a predictor, we correlated the assessment score composites with the actual incidents. As shown in Figure 4, we found a number of strong, negative correlations, such that higher assessment scores were related to fewer numbers of incidents.
Because of the small sample size, these correlations were not statistically significant. However, all of the correlations were in the expected direction and were of moderate to strong magnitude.
To illustrate further the relationship between the leadership assessment and safety incidents, we looked at the number of competency areas for which the managers fell below the footprint minimum score and the total number of incidents. Depending on how an organization decides to use the assessment results, falling below the minimum on a competency is not necessarily a knock-out factor. However, most organizations become concerned when individuals fall below the minimums on multiple areas. The results, shown in Figure 5, clearly indicate that such concerns are warranted.
As Figure 5 shows, managers who fell below the minimum competency cutoffs on three or more competencies had substantially more safety incidents than those without such weaknesses. Although this was a small sample, the results are striking. If this organization used this assessment and screened out individuals who miss four-plus areas (about 33 percent of the sample), it would have 60 percent fewer safety incidents on a yearly basis.
In both of the leadership studies, these managers were from organizations that aggressively committed to safety performance and had instituted behavioral safety programs. Thus, even within groups that institute large-scale organizational development interventions focused on improving safety, some managers clearly do a better job than others. By instituting a job-related and validated selection process, it is possible to improve not only the quality of hires, but also to reduce safety incidents significantly.
Better selection processes in and of themselves are not enough to eliminate accidents and improve safety. However, interventions targeted at improving safety climate and safety performance can be enhanced by selecting people who are less likely to be involved in accidents and more likely to support safety initiatives. Moreover, safety initiatives can be enhanced further by hiring leaders who have better leadership and decision-making skills and are better able to promote and reinforce safety.
Matthew S. O’Connell is executive vice president and co-founder of Select International Inc. He is the author or co-author of more than 100 articles or book chapters, including the best-selling book, Hiring Great People. O’Connell is an adjunct professor of psychology at San Diego State University and director of research and development at Select International. He received his M.A. and Ph.D. in industrial/organizational psychology from the University of Akron. He has designed selection and assessment systems for more than 400 organizations worldwide, including GE, Toyota, Verizon Wireless and the United Nations.
Kristin Delgado is a research consultant at Select International Inc. She received her bachelor’s degree in psychology at the University of Dayton and her master’s degree in industrial/organizational psychology from Wright State University. She is working toward her doctoral degree at Wright State.