PT7.S1.Q10 - a large group of hyperactive children

skrishnan

I'd love an explanation on this one!

Thanks in advance

BinghamtonDave

Can you describe for me what about this question you are having difficulty (if any) getting a handle on? I believe this conversation would be fruitful if I knew your thinking process on this question, that way we can isolate the strengths or weaknesses of your approach and apply those strategies to future LR questions.

David

skrishnan

Hi David,

1.A large group of hyperactive children with high additive diets were observed - let's say 100.
2. These 100 kids were placed on a low additive diet and then observed.
3. Before low additive diet - 60 kids had behavioural issues, ie 40 kids on high additive diets did not have behavioural issues.
4. After low additive diets- 30 kids had behavioural issues.

Now am I wrong in assuming all 100 kids were put on a low additive diet?

conclusion- food additives can contribute to behavioural problems in hyper active children.

I am mainly confused about answer choices B and D.
The reason i crossed out B, is because it's talking about other changes, and in my mind I was like other changes aren't relevant, we are talking specifically about food additives and behavioural problems.

Thanks

BinghamtonDave

Solid write up. Like many problems we encounter on the LSAT, we often have a grasp on the majority of the information, there is simply a gap in our understanding. Fill in enough of these gaps with a concrete understanding of the fundamentals and we are on our way to eliminating mistakes. Points 1,2,3,4, are all an accurate outline of the stimulus in my estimation. At bottom, when we combine your points 1-4 we have correlation in the premise and causation in the conclusion. We observed a correlation and drew a causal conclusion. When we draw a causal conclusion, what we are saying explicitly is: the cause is X. Implicitly, what we are saying is: the cause is not a,b,c,d,e…

So, that is the flaw. The argument moved from correlation in the premise to causation in the conclusion. Now, this question is different because the test writers decided to dig into a reason why the above is actually a flaw. The test writers will do this sometimes with cookie-cutter flaws. They will have a cookie cutter flaw in the stimulus, as clear as day, but there won’t be a cookie cutter description of that flaw in the answer choices. Instead, what there will be is an answer choice that describes why that flaw is indeed a flaw.

Answer choice (B) points out that our study (the correlation) didn’t have a crucial component used to avoid the correlation/causal error in reasoning: a control group. Answer choice (B) says: hey look, you only studied this one group that was placed on the low additive diet. How do you know that there wasn’t another cause? An overarching cause. We would know more about the relationship between the low additive diets and behavioral problems if we had a group who didn’t get the change in diet. Did their behavior changed over the course of the study? That question would go a long way to help us to our conclusion. Instead, we didn’t have that.

Here is a link to an informative video on the importance of a control group.

1:00 starts what (B) is driving at.

An example that points out the flaw in the question. Without a control group, how do we know that every single child had not been subject to a public relations campaign to lower hyperactive behavior by offering toys and video games (bribes) to them? The example is obviously spurious, but without a control group, how can we really isolate that the change observed is due to the cause purported?

David

skrishnan

Wow. I really wasn't looking at this question this way!
This was beyond helpful.

Thanks David!

akistotle

Great explanation as always by @BinghamtonDave!

I learned about the importance of a control group by doing Flaw, Weakening, and Strengthening questions. It is very cookie-cutter.

The list is not exhaustive, but here are some questions that tests your understanding of the importance of a control group:
PT4.S1.Q4 (this is what J.Y. often refers to as "the Dramamine question"), PT25.S4.Q24, PT55.S3.Q9, PT57.S3.Q18, PT73.S4.Q3

skrishnan

@akistotle Thanks for this. I'm currently drilling Flaw. Did a few Pt's and realised drilling will probably be more helpful at this stage.
It really helps you recognise what the question is asking for a lot faster.

akistotle

@skrishnan said:
@akistotle Thanks for this. I'm currently drilling Flaw. Did a few Pt's and realised drilling will probably be more helpful at this stage.
It really helps you recognise what the question is asking for a lot faster.

Yes, drilling is good! Many of the Flaw questions are similar, and it really helps if you memorize certain cookie-cutter molds.

Also, as the video @BinghamtonDave posted explains, it is important to keep in mind that these kinds of studies are only concerned about the average in each group (experimental group and control group). Let's say there is a stimulus that says "since A is correlated X, A is causing X." In Weakening questions, there are (recurring) trap answer choices that say "Some people who did not have A (the control group) experienced an increase in X". But it does not weaken because it is completely normal to have variation. We only care about the average in each group.

Good luck

skrishnan

That's great input!
Thank you so much @akistotle
You guys are great