Skip to content

Thinking and Deciding 8: Judgment of correlation and contingency


Attentional Bias

This is the most common error when judging correlation. There are four relevant pieces of information when judging correlation of X and Y, and they include X present Y present, X present Y absent. X absent Y present. X absent Y absent. All four pieces are important, but people tend to look only for what the case is when X is present. If Y is often or usually present along with Y, then people will think they are correlated. But if it is the case that even when X is absent, Y is present just as much, then there is no correlation. This is the “count the hits, skip the misses” of prayer answering and psychic phone predictions.

Effects of prior belief

When data exists on a topic, people’s preconceived conclusions color how they view the data. In psychological tests like the Rorschach, subjects based their judgments of correlation on their previous beliefs, when such correlations did not exist (and other correlations did in fact exist).

Judgment of personality traits that co-occur was also flavored by what “ought” to be true. For example, in one experiment there was little correlation at a summer camp between spontaneous outbursts of joy or disapproval and talking too much at the dinner table. But when counselors were asked to rate how much these traits went together, they based their memory of the events on their prejudices, not their observations.

Attentional bias comes into play in prior belief as well. People tend to ignore data against their beliefs, and seek out data for their beliefs.

Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: