Research question

[basophilic]

Question: A new experimental correctional facility for boys has been established based upon the principles of holistic healing. In accordance with the hypothesis that a diet high in sugar can exacerbate violent tendencies in young people, access to sugary foods is wholly restricted in this residential facility. After the first six months, the data indicate an average 27% reduction in violence compared to pre-admission levels (assessed upon intake). Nearly all of the boys had diets containing substantial amounts of sugar prior to admission. The chief psychologist then publishes his finding that these data indicate that reduced consumption of foods high in sugar may lead to significant violence reduction in youths. Which of the following characterizes the greatest problem with that finding?
A) The study failed to consider or control for the fact that foods high in sugar (i.e. “junk food”) may also tend to contain high levels of other potentially problematic ingredients.
B) A 27% reduction in violence is not substantial enough to justify proposing a connection between the two variables in this case.
C) All of the subjects in the study were male.
D) The study lacked a control group.

Answer is D. I have a few questions:
1) Doesn't the wording in the penultimate statement in question stem imply that the psychologist sees a potential causation? TPR says it implies only a correlation.
2) What makes D better than A? A mentions "control" as well (albeit a more specific one) while also accounting for confounding variables.

---

Members don't see this ad.

 

[bee17]

All answer options are potential problems. However, the biggest problem with this study is that there is no control group, which means no conclusions can be drawn from this study. The study should have included another group of boys in a correctional facility that were given a typical diet for sake of comparison. What if the reduction in violence was due to being in a correctional facility rather than diet? The other potential problems (different ingredients, levels of violence reduction, and gender bias) cannot be addressed until the fundamental problem of a lack of a control group is addressed.

 

[basophilic]

All answer options are potential problems. However, the biggest problem with this study is that there is no control group, which means no conclusions can be drawn from this study. The study should have included another group of boys in a correctional facility that were given a typical diet for sake of comparison. What if the reduction in violence was due to being in a correctional facility rather than diet? The other potential problems (different ingredients, levels of violence reduction, and gender bias) cannot be addressed until the fundamental problem of a lack of a control group is addressed.

Ah ok. So in general, for most experiments you need to have a control group BEFORE you can consider potential confounding variables?

 

[bee17]

That would be a good rule of thumb.