Research is (not) trivial

“Churchgoers have more children than people who stay away from church.” This finding might seem trivial to many people. Everybody can think up one or two reasons for the fact. Nobody needed research for this.

Social sciences are particularly susceptible to the accusation that they only say what common sense has long told us. But is this the case?

Many studies have looked into this topic (e.g. Baratz, 1983; Slovic & Fischhoff, 1977). Typically, one group of people is shown real study results (e.g. on the link between faith and parentage). These people (students or experts) are asked to rate whether they find the data trivial, uninteresting, or not very surprising. Another group is given the exact opposite, fictivious data. They are also asked to rate how trivial the data is. The surprising result: Both groups find it just as trivial!

This “I knew it all along” effect (Wood, 1978) is explained with the hindsight bias and our excessive trust in our own judgment (Slovic & Fischhoff, 1977). Looking back, we overestimate how easily an event could have been predicted. People even systematically mis-remember their original assumptions.

Thus, accusations of triviality of research results should be used sparingly. The same goes for the findings of staff surveys (Borg, 2003). Managers would do well to sit down and think what they expect before the results come back. Then, they will see (and can use) the information that is truly surprising.

Sources