Make people follow you for the right reasons

Why is it that the perception is you're either a "numbers geek" or a dynamic leader? Isn't there room to be both?

"The definition of leadership is knowing when to react and when not to react," Linafleter said. "Physicians, for example, are trained to make quick decisions. They gather the data, make the diagnosis, and decide on the treatment. Their basic course of action can be described as, 'Problem? Fix it. Problem? Fix it.'

Linafelter cited the work of W. Edwards Deming, the statistician and consultant widely recognized for helping turn Japan into a post-World War II economic powerhouse by teaching statistical process control methods to managers to help them improve design, service, and product quality.

One of Deming's creations was the control chart, which tracks performance of a certain task over time. As might be expected, any task in which humans are involved will include some slight variations in quality. Those "up and down" variations do not signal the need for management to react. Sudden, big changes or "spikes" in performance, however, do signify the need for action.

"This is a good example of how being a 'numbers geek' can help you be a better manager, of how the two concepts are not mutually exclusive," Linafelter said. "Look at med errors, for example. No one comes into work thinking he or she is going to make med errors that day. Yet when you go to any hospital and measure, you'll find that med errors occur virtually every day.

"This is where being a number-cruncher comes into play: the ability to know when to react. Knee-jerk reactions to variations from 'common causes' (such as humans) can be very dangerous, as can not reacting because you're failing to recognize when variations are caused by an outside force (one that causes a spike)."

Linafleter said that, in addition to medication errors, control charts are valuable tools for tracking many variables in a health-care setting, such as staff turnover and patient satisfaction.

"Control charts are the essence of what you do as a manager," he added. "You must decide when and how to react, because the manager who reacts to everything is never going to get anything done, and the manager who reacts to nothing isn't a manager."

Although statistical analysis can be a valuable management tool, Linafleter said, its primary use is to identify and calculate the limits of "standard deviation," not to establish "goals."

"The true goal is to reduce the deviation by improving the process. Just because a process is under control doesn't mean it's the right process," he said.