Percent safe behavior

Posting provides everyone with knowledge of results. No one knows what the percent safe behavior is in various areas without behavioral sampling, tracking, and posting. This awareness of risky behavior and trends is a beginning step toward improvement. 

We might be reminded of a general purpose—"Actively Care for a Total Safety Culture," or challenged—"100 Percent Safe Behavior is Our Goal This Year."... 

Feedback is essential. Each observation process with a CBC provides for tallying and graphing results as "percent safe behavior" on a group feedback chart The CBC shown... 

As an example, in evaluating the safe behavior of 1000 men and women. Table 12.1 shows the percent of the total that either did or did not have occupational injuries. The contributing events (men having injuries and women having injuries) are not mutually exclusive since either or both may have suffered occupational injuries and, subsequently, affected the primary or top event (safe behavior). Hgure 12.7 shows the fault tree for this extremely simple example. By applying the modified formula for non-mutually exclusive events, the probability of an injury event I involving a man M can be calculated as follows ... 

The headways from the three phases of Ihe drive are presented in Figure 5-6. Even with the instruction to avoid headways less than one second, the typical headways were 0.4-0.8 seconds, and when drivers followed another vehicle (before permitted to pass), nearly half die time (42.2%) they maintained headways of less than 0.8 seconds. In contrast to this dangerous pattern, the introduction of feedback caused a dramatic shift in car following behavior. The percent of time with headways less than 0.8 seconds dropped from over 40 percent to to 3.5 percent. Even after the feedback was removed, during the final 20 kilometers, the drivers maintained that safe behavior and drove with headways less than 0.8 seconds only 6.5 percent of the time. Thus, in the process of receiving feedback, the drivers must have internalized some critical cues that enabled them to judge 1 second headways fairly accurately, so that they performed nearly as well once the external feedback was removed. 

These data are not meant to suggest that employees are directly to blame for 96 percent of their injuries. From the perspective of behavioral psychology, all behavior is a function of the environment in which it occurs. Unsafe work behavior is accordingly the result of (1) the physical environment, (2) the social environment, and (3) workers experience within these. The remainder of this book is dedicated to how to change the workplace environment in ways that increase safe behavior. 

This is not to suggest that managers and supervisors should forego attention to safety on the job. The observation data are a measure of the extent to which employees work safely. Managers and supervisors must provide daily feedback to employees for on-the-job safety to ensure the success of the behavioral safety process. However, they should emphasize safety on the job and maintaining scheduled observations, not the percent safe resulting from completed observations. 

During performance reviews wifli managers and supervisors a) Assess level of support and participation in behavioral safety process (percent of completed observations, not the percent safe resulting from observations)... 

As to posting the data, many of a company s considerations with peer observation data pertain also to the self-observation process. The data should be posted in the respective group s or crew s work area for review only and updated weekly. The applicable safety representative incorporates discussion of trends in the current data, analyzes specific behaviors that are getting many no s, and helps determine when changes or new behavior items should be incorporated on the index. If the scores begin to achieve 90 to 100 percent safe for five or six samplings, this usually indicates that the behaviors listed have started to get to a high and steady rate. It would make sense at this point to develop a new index with safety behaviors that help address new concerns that arise in the increased awareness stimulated by the... 

Step 5 Select the target practice. Combine the information from steps 1 to 4 to select a behavior to target for improvement based on the probability of injury (based on the number of injuries that have resulted), the severity of potential injury, exposure (the number of times employees perform the practice), consistency (based on the percent safe on a particular concern), and trends in your data. 

Feedback Support Cards. Written cards can be an effective way to support verbal feedback as part of the observation process. They may be thank-you cards distributed to employees who are performing safely on the day they are observed. Observers may give the cards to employees to support a specific safe behavior that is being emphasized by a special campaign, such a proper lifting in conjunction with a back safety campaign. Or the observers may give the cards to everyone in an area with 100 percent safety performance. 

I told Krista her "percent safe" score and proceeded to show her the list of safe checkmarks, while covering the checks in the At-Risk column. Obviously, I wanted to make this a positive experience, and to do this, it was necessary to emphasize the behaviors I saw her do correctly. To my surprise, she did not seem impressed with her 85 percent safe score and... 

The formula at the bottom of the CBC in Figure 8.12 can be used to calculate an overall percent safe score. We have found it very effective to post this global score weekly for different work teams. Such social comparison information presumably motivated performance improvement through friendly intergroup competition (Williams and Geller, 2000). Chapter 12 also includes addition information on e design of CBCs for one-on-one behavior observation. 

The checks in the safe and at-risk columns of a CBC can be readily summarized in a "percent safe score." As I discussed in the previous chapter, an overall global score can be calculated by dividing the total number of behavioral observations (i.e., all checks on all CBCs) into the total number of safe observations (i.e., all checks in the safe columns of all CBCs). This provides an overall estimate of the safety of the workforce with regard to the critical behaviors targeted in the observation step of DO IT. 

The global "percent safe score" does not provide direction regarding which particular behaviors need improvement, but it can provide motivation to a workforce that wants to improve (Williams and Geller, 2000). It is an achievement-oriented index that holds employees accountable for things they can control. This assumes, of course, that the workers know the safe operating procedures for every work task. 

If some employees are not sure of the safe way to perform a certain job, behavioral direction is needed. A global "percent safe score" is not sufficient. When the CBC is reviewed during a one-on-one coaching session, behavioral direction is provided. The worker sees what critical behaviors were observed as "safe" and "at-risk." A constructive conversation with the coach provides support for safe behavior and corrective feedback for behavior that could be safer. Often this includes suggestions for making the safe behavior more convenient, comfortable, and easier to remember. It might also include tire removal of barriers (physical and social) that inhibit safe behavior. 

Groups can receive support and direction for specific behavior if a "percent safe score" is derived per behavior. That is, instead of adding safe and at-risk checks across behaviors, calculate separate percentages for each behavior. Then the group can see which critical behaviors are safe and which need improvement. The results can be posted in a prominent location, as depicted in Figure 9.1, and discussed in team meetings. 

Because global feedback was just as effective as specific feedback when social comparison feedback was included, we presumed most of the 97 employees of the soft-drink bottling facility did not need an instructional intervention. They knew how to perform their jobs safely but needed some extrinsic motivation to follow the nine safety policies implied by the nine target behaviors. This was provided by a global percent safe score from a similar work group. 

Coaching was presented as a rather formal step-by-step process whereby a critical behavior checklist (CBC) is developed and used to observe and analyze the safe vs. at-risk behaviors occurring in a particular work procedure. Then, as illustrated in Figure 13.1, the CBC is used to present directive and/or motivational feedback in a one-to-one interpersonal conversation. Also, percent safe scores are derived from a variety of CBCs and presented on a group feedback chart. Comments written on the CBCs are discussed in group meetings to analyze areas of concern and to find ways to make safe behavior more likely to occur. 

As one can see from the above discussion, there can be many causes of unsafe conditions, but, usually unsafe conditions are not the sole cause of an injury. It is believed by many that unsafe behavior accounts for 80 to 90 percent of all injuries. Some contend that OSHA s focus on things, such as machine guards, handrails, and protective equipment, will prevent only 20 percent of all injuries. However, safe behavior is mostly management s responsibility, not OSHA s. Therefore, management must focus more on a comprehensive safety program just abiding by OSHA regulations is not the answer. 

While the manufacturers of measurement devices can supply some information on the dynamic characteristics of their devices, interpretation is often difficult. Measurement device dynamics are quoted on varying bases, such as rise time, time to 63 percent response, settling time, and so on. Even where the time to 63 percent response is quoted, it might not be safe to assume that the measurement device exhibits first-order behavior. 

---