Correlational Data is NOT CAUSAL...

I read an article at the following link

http://www.healthsentinel.com/org_news.php?id=075&title=Parental+stress+a+factor+in+ADHD+diagnosis&event=org_news_print_list_item

I always have a problem with the way correlational data is interpreted and reported in scientific journals and in the press. Correlational data does not imply cause and effect...yet even though all psychologists know this, the discussion section of journal articles almost always draws a causal conclusion (e.g., Parenting stress is related to reported ADHD symptoms in children...therefore, stressed out parents must over-report symptoms or create ADHD symptoms in their children due to poor parenting practices). Even though this is likely true, the writers ignore the other possible explanations. There could be many other variables that account for parents' reporting of ADHD symptoms, or for parents being stressed. For example, children that are hyperactive and inattentive are very stressful to raise and require constant supervision. The children with ADHD could be causing the stress in the parents. There could be something else happening at home that leads to ADHD-like symptoms, such as low blood sugar, tiredness in the evenings after school, increased conflict with siblings that does not necessarily happen at school, etc. Just because parent and teacher reports don't usually agree, doesn't mean the symptoms are overreported, or that the parents are causing the symptoms. Correlational data does not allow us to infer causal links.

I just re-read the article and I guess these authors were not blatantly saying parenting stress was causal in nature, but only that it is an important variable to consider when doing an ADHD assessment...but, I do see the causal links made all the time in articles and it is bothersome to me.