The “D” of the LEAD acronym I introduced in my ISHN column this past December stands for “Data” (see sidebar). Data both directs and motivates behavior. By observing the results of our actions, we learn how well we completed a task and what we can do to improve.
But some data are useless, misleading, and de-motivating. For example, injury statistics based on self-reporting are unreliable and have no diagnostic value. And they can activate distress or a false sense of security. Leaders need to use data strategically to direct and motivate themselves and others.
“What gets measured gets done.” This popular slogan reflects the connection between data and accountability. But using wrong data to assess accountability can be disastrous. “What could be worse,” asked Dr. Edwards Deming, than “holding willing workers accountable for numbers they cannot control?”
Dr. Deming taught us the critical difference between behavior and performance, a distinction needed to select and examine the right data. Many behavioral researchers and safety professionals use these words interchangeably, but my online dictionary (www.m-w.com
) defines performance as “something accomplished” and behavior as “the manner of conducting oneself.”
This behavior/performance distinction is critical for giving the right kind of feedback. Specifically, when can we hold people accountable for data? The answer is simple. Hold people accountable for data they directly influence.
In safety, it’s fair to hold ourselves accountable for the variety of activities we can do to prevent personal injuries â€” from coaching others regarding their safe versus at-risk behaviors to completing hazard recognition and close-call reports. Likewise, if an individual’s behavior or lack thereof is clearly linked to an injury, it is legitimate to hold that person accountable (in part) for the performance data reflected by injury statistics. But the contribution of environmental factors beyond the individual’s control should be acknowledged.
Some performance deficits result from behavior deviating from the process. But performance deficits also occur from system factors (physical conditions, management decisions) independent of process-related behavior. Hold people accountable for the first, but not the latter.
Isn’t this common sense? Then why does there seem to be so much emphasis on injury statistics or performance data at safety meetings? How often is a graph of safety-related behavior displayed to illustrate accomplishment (or failure) at injury prevention? Bottom line: Show process data to individuals and groups that reflect their controllable actions associated directly with performance data.
Almost every book on leadership presents information on the person characteristics of leaders. For example, the recent text by Dr. Thomas Krause, “Leading with Safety” (John Wiley & Sons, 2005), connects leadership with five personality traits â€” emotional resilience, extraversion, learning orientation, collegiality, and conscientiousness. Krause also distinguishes between transactional leaders (or managers) and transformational leaders with certain interpersonal styles (including challenging, engaging, inspiring, and influential). In an earlier ISHN article on leadership (March 1995), I described leaders as individuals who are energetic, passionate, open, trustworthy, compassionate, goal-directed, self-confident, intelligent, and flexible.
It’s fascinating and entertaining to explore one’s personality, and consider correlations between specific person factors and behavior. Many readers have taken the Myers-Briggs personality inventory, and enjoyed learning about the behavioral implications of certain person qualities and styles.
I urge caution when considering these data. First, the assessment tools for personal data are often unreliable and invalid (as covered in my ISHN articles for November and December 1994). Secondly, the connection between most person data and behavior is ambiguous or weak. But the critical issue is applicability.
Can data suggesting leadership-related personality traits, states, or styles provide directional or motivational feedback to an individual? Actually, using these data to influence ourselves or others is analogous to developing an action plan from an organization’s injury data. In both cases, the data are unreliable and influenced by undefined factors independent of people’s behavior. And neither provides useful diagnostic information to direct continuous improvement.
Dr. Krause acknowledges low practical value in assessing leadership-related characteristics of people. Telling people they score high or low on a measure of charisma gives minimal direction for improving leadership. But to the extent it’s possible to define a particular leadership quality in terms of specific behaviors, personality data can be useful. For example, by observing people judged to be charismatic, it might be possible to identify behaviors that reflect this label and then tell people what they can do to demonstrate charisma. Subsequently, a person can be observed and given behavior-based feedback related to the presence or absence of charisma-related behaviors.
Aubrey and Jamie Daniels advance an entirely different perspective in their book, “Measure of a Leader” (Performance Management Publications, 2005). They claim the measure of a leader should focus on the behavior of the followers. The key type of follower behavior: “discretionary behavior” supporting the leader’s vision.
This is behavior that exceeds a worker’s job requirements. It is self-directed, meaningful, and intrinsically reinforcing. I refer to this type of behavior as “actively caring” whenever it relates to injury prevention or health promotion.
Increasing discretionary behavior
The Daniels brothers focus on the appropriate use of “positive reinforcement” to increase discretionary behavior. With threats and punitive consequences, people do not become self-accountable, and do only what’s required.
The approach I’ve advocated for increasing actively-caring behavior is consistent with these suggestions. Research indicates people are more likely to help others (or emit discretionary behavior) when they have relatively high levels of self-esteem, self-efficacy, personal control, optimism, and a sense of belongingness.
Genuine behavior-based rewards and recognition are likely to enhance these person states. But there are other ways to facilitate the occurrence of these person states and increase the probability of discretionary behavior.
One final point: Please be skeptical of people’s opinions, even if they sound like good common sense. I recommend frequent use of the phrase, “Where’s the data?” And when someone shows you data, ask another question, “How can these data be used to facilitate continuous improvement?”
SIDEBAR: Leadership principlesL
for Live, Listen, Learn, Love, and Leave a LegacyE
for Empathy, Energy, Empowerment, and EngagementA
for Audacity, and Achievement of success over Avoidance of failureD
for Data to support an opinion or perspective