A study by Caspi and others (1995) revealed that temperamental style in early childhood might be linked to adjustment problems during adolescence. The study involved a cohort of 1,037 children born between April 1, 1972, and March 31, 1973, in Dunedin, New Zealand.

At ages 3, 5, 7, and 9, the behavior of each child was rated in terms of 22 aspects of temperament, including emotional stability, restlessness, self-reliance, persistence, negativism, passivity, shyness, self-confidence, emotional flatness, and 13 other dimensions. To assess behavior problems in the sample, the researchers relied on outcome data from teachers and parents who rated the children at ages 9, 11, 13, and 15 on two widely used behavior problem checklists. These checklists included subscales measuring anxiety/withdrawal, which represents feelings of inferiority and failure; attention problems, which reflect difficulty in concentration skills; conduct disorder, which reflects aggressiveness and alienation; and socialized delinquency, which reflects norm-violating tendencies.

For both boys and girls, lack of control at ages 3 and 5 showed a significant positive correlation with teacher and parent reports of antisocial behavior and conduct disorder at ages 9 and 11. In addition, boys and girls characterized as lacking in control in early childhood were less likely to be rated in adolescence as mature and confident.

The authors suggest several possible explanations for these intriguing results. One is that certain temperamental characteristics in young children are actually early, "subclinical" manifestations of more extreme behavior disorders. Whatever the correct explanation, the results of this extensive study suggest that early temperament may have remarkably specific predictive validity for the development of behavior problems during adolescence.

1. What type of research design is used in this study (e.g., cross-sectional, longitudinal, experimental, correlational, or naturalistic observation)? Is this design appropriate?

2. What explanation do the researchers offer for their findings? Does this explanation make sense based on the evidence?

3. Given the results of this study, why can't the researchers draw a causal connection between behavior disorders and temperament?

4. Can you think of an alternative explanation for the results of this study?