Many researchers are interested in the way that childhood interactions with other human beings influence child development. For example, a great deal of research suggests that children who receive more attention from adults tend to have better social and cognitive development outcomes than those who receive less.
This research, however, is entirely observational and the estimates of the impact of adult attention may be biased by the fact that the child’s personality itself may be part of the reason that adults either do or do not pay attention to him. For example, a precocious and charming child may receive extra attention from caregivers and teachers simply because the adults enjoy interacting with that child. Those same qualities, intelligence and charm, are also likely to be correlated with good cognitive and social outcomes. Since the child’s personality in this scenario is influencing both the dependent and independent variables of interest, it becomes a source of potential omitted variable bias.
Estimates of the impact of adult interaction on child development in this scenario would be biased by the fact that the estimates would be “soaking up” the impact of the personality traits that cause adults to pay attention to a particular child rather than others.