A study of a new medication found that patients who took the medicine have reduced blood pressure, compared to those who did not take the medicine. The research team concluded that the medicine caused the reduction in blood pressure. Which study design did they most likely use?

Question

animusleviosaa · Answer

They most likely used an experiment with patients randomly assigned to take the medicine or not take the medicine. <-- what others think
 They most likely used an experiment with patients randomly assigned to groups based on their blood pressure.
 They most likely used a survey asking patients if they took the medicine and how their blood pressure was affected.  <-- what I think
 They most likely used a survey asking patients if they took the medicine and if they felt the medication helped their blood pressure.

animusleviosaa · Answer

@jim_thompson5910 would you mind helping in one more?

jim_thompson5910 · Answer

`A study of a new medication found that patients who took the medicine have reduced blood pressure, compared to those who did not take the medicine`

we have 2 groups
Group A = group that took the medication = experimental group
Group B = group that took the placebo = control group

this shows us that they're doing an experiment and NOT a survey. The scientists are holding everything else constant and applying treatment to see if the medicine works or not

jim_thompson5910 · Answer

a survey would be if the scientists were just observing and not altering or changing anything they are studying

eg: taking a survey to see people's views on a particular political topic

animusleviosaa · Answer

Okay so it would be the first option? Thank you for explaining, the reasoning just sounded better on the survey ones

jim_thompson5910 · Answer

yes the answer is Choice A) They most likely used an experiment with patients randomly assigned to take the medicine or not take the medicine.

animusleviosaa · Answer

Thanks again

jim_thompson5910 · Answer

sure thing

mww · Answer

Also helpful to understand only experiments can conclusively imply 'causation' between two variables. If this is just an observation of events, then we can only measure whether there is a correlation between the two variables.

For example it could be that the people taking the medicine were also on a special exercise regime which could affect the outcome observed. If you did not control for this condition as you would in an experimental design, you would incorrectly assume the medicine was the cause of the change when it could have been the exercise regime. You could say the medicine was associated with the change however (and may also find the exercise to have an effect, after which you can design and experiment to test which factor(s) are significant).