Selection bias

If you’re a coach or trainer, you likely will encounter many people who are extremely nervous about public speaking, and want to be less nervous. Also, you will encounter non-nervous people who greatly enjoy speaking, and seek to improve from good to great. You will not encounter slightly nervous people, who won’t bother to ask for help. Based on that nonrandom sample from the general population, you may be tempted to incorrectly conclude that most people have a phobia about public speaking. (Only about 10% of U.S. adults have a phobia). Selection bias is a problem for all professionals, and it will distort how you see the world.

When I was a medic in the Air Force Reserve in the early 1970s, we did routine flight physicals, each of which included a 12-lead electrocardiogram (EKG). We used a special die to trim 3” samples from those 12 strip charts, and mounted each set on a self-adhesive form. A wise old man, Dr. Richard V. Lynch, was commander of our clinic at the Greater Pittsburgh Airport.

Each afternoon Colonel Lynch would eagerly pounce on that stack of EKG forms and go through them with a fine-tooth comb. He even replotted them as vectorcardiograms. Occasionally he’d frown, mark one of the forms “apparent extreme bradycardia” and tell us to please rerun it. That meant we’d accidentally flipped the mode lever on the machine one notch too far, and ran it at double speed (50 mm/sec) rather than normal speed (25 mm/sec).

Dr. Lynch also was professor of cardiology at the West Virginia University medical school. He had realized that if a specialist doesn’t do something about it, he mainly will see referrals where another physician first looked at the patient, ran an EKG, and noticed there was something very wrong. His diagnostic skills gradually will deteriorate. He knew the way to avoid that problem was by looking at a sample of healthy people every month. Those flight physicals were his reality check for staying sharp.