tag:blogger.com,1999:blog-521714752439066565.post2307162907791098913..comments2009-11-05T08:45:12.531-08:00Comments on Alan & Bo's Correlation & Causality Blog: Child Development Article on Spanking and Early Child Developmentposted by the staffhttp://www.blogger.com/profile/13525349478448118800noreply@blogger.comBlogger1125tag:blogger.com,1999:blog-521714752439066565.post-48167944488488586742009-11-05T07:50:23.567-08:002009-11-05T07:50:23.567-08:00Alan,
Thank you for a very thoughtful analysis of...Alan,<br /><br />Thank you for a very thoughtful analysis of this research. Much appreciated. But I still see uncontrolled factors that seem quite plausible as third variables that would not even permit a mildly causal inference. For example, it does not appear that the researchers controlled for the violent tendencies of the parents. And I am not even talking about the passive gene-environment correlation you discuss. I am simply considering the environmental influence of a more aggressive parent. Couldn't that account for more spanking at time 1 and greater child aggression at time 2? As you note, there are other significant psychopathology, mild or otherwise, that could influence the variables.<br /><br />Also, I think the use of "can" deserves discussion in the presentation of scientific findings. What does it really mean? Often it appears to be presented as a qualifier to the claim presented. That is, the author of the statement means to suggest that there is some evidence to support the claim (e.g., causal), but the evidence is not that strong or conclusive yet. But the author is still making the (causal) claim.<br /><br />Or, I see "can" used to simply represent speculation. We don't really have evidence to support a claim, but perhaps it is true. We only have a correlation at this point, but perhaps there is a causal relationship between the two variables.<br /><br />I speculate that the majority of the public would interpret the claim that "A can have negative consequences for B" to mean that we have some evidence that at least in some cases A does affect B.<br /><br />How should we use words "can" or "may" or "it appears" when we present scientific claims?<br /><br />Jon MuellerJon Muellernoreply@blogger.com