This Article  • Full text  • Full text PDF  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Contact me when this article is cited  Related Content  • Similar articles in this journal

Arch Pediatr Adolesc Med. 1998;152:305-309.

	Since this article does not have an abstract, we have provided the first 150 words of the full text and any section headings.

Two recent articles published in the ARCHIVES^1-2 argue that they have found evidence for "causal" relationships between spanking and antisocial behavior in children, such that increased spanking causes antisocial behavior. Unfortunately, their methods do not allow for such conclusions. In fact, their methods do not allow for any conclusions at all. I believe it is particularly important to point out these mistakes because they have become commonplace in the social sciences³ and it is important that these mistakes do not become commonplace in medical research.

One initial mistake made by both authors is the claim that they are testing for causality with longitudinal data.¹ Causal inferences can only be drawn from experiments.⁴ What can be tested for with longitudinal data is temporal order.⁵ Temporal order is frequently cited as an important aspect of causality.⁴

In longitudinal research, the temporal order between variables can be known or unknown. For example, the . . . [Full Text of this Article]