Scientific Reasoning

# Unit #7 Summary

In this unit we looked at some forms of reasoning commonly used in the sciences. We began by looking at some important components of statistics and their visual representations. In Chapter 20 we turned to an examination of causal language, and a particularly important kind of comparative statistic: the correlation. Although correlations can be especially useful starting point for identifying likely cause/effect relationships, not every correlation captures a causal relation. Before drawing an inference like this, we need to check to see whether other conditions have been met. Chief among them is whether there are other possible explanations for the correlation besides the hypothesized causal relation. As we saw, there are all kinds of techniques for identifying significant correlations, though some techniques are better than others.

A Key Question for a Statistic:

- What, exactly, was counted?

Questions to Ask ofInferences from correlation to cause:

- How likely is the proposed explanation?
- Are there other plausible explanations for the correlation?
- Would the truth of the proposed explanation be less surprising than the truth of any competitor?

## Key Terms:

- Statistical Claim
- Comparative Statistics
- Outliers
- Scale
- Relevantly Similar Statistics
- Causal Inference
- Complete Cause
- Partial Cause
- Reliable Cause
- Probabilistic Cause
- Positive Correlation
- Negative Correlation
- Significant Correlation
- Observational Study
- Experimental Study
- Experimental Group
- Control/Control Group
- Confounding Factor
- Controlling for a Variable
- Unknown Confounders
- Experimental Group
- Control Group

## Further Reading

There are a number of short readable books about the pitfalls of dealing with statistics. A famous, but dated, source is *How to Lie with Statistics* by Darrell Huff and Irving Geis. For more contemporary examples, you could check out Joel Best’s *Damned Lies and Statistics* or *Is that a Fact? A Field Guide to Statistical and Scientific Information* by Mark Battersby. For a deeper discussion of experimental structure and set up, see *Randomized Controlled Trials: Questions, Answers, and Musings* by Alejandro Jadad and Murray Enkin.