Building One's "Scientific Intuition"

[DendWrite]

---

Members do not see this ad.

A lot of my exams recently have been focused on interpreting experimental data in the context of the general concepts that we've learned from class. I have general knowledge about how to interpret data (i.e. how to read graphs and things), but have any of you found ways to improve your skills at interpreting / extrapolating from data on either experimental questions on tests or passages like this on practice MCATs? Do you know of any good books that deal with this topic? I know that I could go to PubMed and just read tons and tons of papers, but I was hoping to figure out some more broad strategies rather than a brute-force approach. Any tips?

 

[randombetch]

Read tons of Nature papers - they're interesting and well written.

 

[greatnt249]

A lot of my exams recently have been focused on interpreting experimental data in the context of the general concepts that we've learned from class. I have general knowledge about how to interpret data (i.e. how to read graphs and things), but have any of you found ways to improve your skills at interpreting / extrapolating from data on either experimental questions on tests or passages like this on practice MCATs? Do you know of any good books that deal with this topic? I know that I could go to PubMed and just read tons and tons of papers, but I was hoping to figure out some more broad strategies rather than a brute-force approach. Any tips?

For starters, rarely do you just look at a graph and try to interpret it. It's usually provided along with some contextual passage that explains what the intent of the experiment being performed was and how the results (shown in the graph) support any conclusions from said experiment. Writers depend on the fact that you are aware of protocols and limitations of analytical techniques so that you can understand the case they're trying to make by presenting data from their experiments. So, that being said:

1. Read the passage.
2. Figure out the intent of the experiment.
3. Figure out which techniques were used in the experiment (essential).
4. Know how using those techniques are relevant to the case being made (essential).
5. Note units being used in charts and graphs.
6. Analyze trends in data (positive/negative/no trend whatsoever).
7. Predict what the graph should look like if the results support/negate conclusions drawn from data.
8. See if the results correlate with the case being made.

That being said, it's difficult to know what exactly you mean by "wanting to learn how to interpret graphs better"; is there something in particular you are having trouble with?

 

[Lokhtar]

I second the reading Nature articles.

 

[boaz]

I would add that working in a lab (research, not undergrad intro), if you don't already, is an excellent way to hone that scientific intuition. The work should involve responsibilities that require you to do just that sort of thinking. Try to get involved in the nuts and bolts of the ideas being researched rather than just doing your job. And it doesn't have to be revolutionary research at a world-class institution.