## MCAT Physics and Math Review

**Chapter 12: Data-Based and Statistical Reasoning**

### 12.7 Applying Data

Finally, we have reached the discussion section of an academic paper, in which the data that we have gathered and interpreted is applied to the original problem. We can then begin drawing conclusions and creating new questions based on our results. Because much of this was covered in the discussion on experimental methods in Chapter 11 of *MCAT Physics and Math Review*, we will be terse in our review here.

CORRELATION AND CAUSATION

As discussed previously, we must be careful with our wording when discussing variable relationships. **Correlation** refers to a connection—direct relationship, inverse relationship, or otherwise—between data. Correlation does not necessary imply **causation**; we must avoid this assumption when there is insufficient evidence to draw such a conclusion. If an experiment cannot be performed, we must rely on Hill’s criteria, discussed in Chapter 11 of *MCAT Physics and Math Review*. Remember that the only one of Hill’s criteria that is uniformly necessary for causation is temporality.

IN THE CONTEXT OF SCIENTIFIC KNOWLEDGE

When interpreting data, it is important that we not only state the apparent relationships between data, but also begin to draw connections to other concepts in science and to our background knowledge. At a minimum, the impact of the new data on the existing hypothesis must be considered, although ideally the new data would be integrated into all future investigations on the topic. Additionally, we must develop a plausible rationale for the results. Finally, we must make decisions about our data’s impact on the real world, and determine whether or not our evidence is substantial and impactful enough to necessitate changes in understanding or policy.

**MCAT Concept Check 12.7:**

Before you move on, assess your understanding of the material with these questions.

1. True or False: Statistical significance is sufficient criteria to enact policy change.

2. True or False: Two variables that are causally related will also be correlated with each other.