Yes. And the researchers know that too. There's an entire section of the paper devoted to teasing this apart. From the third paragraph of the paper ("Methods")
"[T]here are multiple threats to the validity of such an approach. We were particularly concerned about factors affecting a young person's probability of attending Scouts–Guides, which could also be protective of mental health in later life. To test for this possibility, we used three approaches. First, we explored a form of ‘negative control’ exposure. Specifically, we investigated whether attendance at other types of organisations (voluntary groups and the church), and other outside school clubs, which might have a similar set of common factors encouraging attendance and which might be protective of mental health in later life, were also associated with better later life mental health, conditioning on scouts–guides attendance. The absence of an association would suggest that an unmeasured common factor was not confounding the scout–guide association, while the existence of an association would be ambiguous with an unmeasured common factor and a third factor, independent of scout–guide attendance, but acting in a pathway through the other types of organisation on mental health, being possible. Second, we sought evidence of a dose–response relationship, exploring whether fractions of attendance were related to better mental health. Third, we explored whether the Scout–Guide association was weaker in regions with a higher proportional attendance which might suggest Scout–Guides not being the causative agent."
I dunno why you're getting down voted as this is a clear point to make. A fun game I like to play when I read these sorts of things in the news is to reverse the causation implied by what I'm reading. Often times it make more sense. In this case:
Those with better mental health are more likely to do Scouts and guides in their youth.
I find such a comment to contribute more to the discussion (if it's actually discussed) than all the anecdotes from many of the other top-level comments. They are nicely worded and afford many points that can be discussed, most of them irrelevant to the matter of the article.
It's not - it points out that the two conclusions have equal supporting evidence, but different implications. Hence, based on the information in the article, there's no reason to expect that on average, a child joining scouts would improve their lifetime mental health.
Conversely, looking at the source journal article, the researchers did actually check for confounders in several ways, giving some credence to the idea of a causative effect of scouts participation on lifetime mental health.
The article was bad, but the study is hard to read. I'm not 100% sure but I think this this blip says scouting potentially has a casual relationship. It shouldn't translate to "Scouts provide mental health boost".
>This study used a prospective cohort design, controlling for many of the factors that might affect participation in Scouts–Guides and later life mental health. Most of these measures were collected in the appropriate cohort sweep. The possibility of residual confounding, particularly due to missing or poorly measured socioeconomic variables, was further examined using three methods, a negative exposure control, an exploration of ‘dose–response’ and a measure of differing regional rates of Scouts–Guide attendance. Although there was no evidence of a dose–response relationship, the other methods supported the plausibility of a causal relationship.
