Summary

• Science is a process of testing statements about how the natural world works—called hypotheses. Scientific hypotheses must be testable and falsifiable. Hypotheses are tested via the process of deductive reasoning, which allows researchers to make specific predictions about expected observations. Absolutely proving hypotheses is impossible. However, well-designed scientific experiments allow researchers to strongly infer that their hypothesis is correct.

• Controlled experiments test hypotheses about the effect of experimental treatments by comparing a randomly assigned experimental group with a control group. Controls are individuals who are treated identically to the experimental group except for application of the treatment. Bias in scientific results can be minimized with double-blind experiments that keep subjects and data collectors unaware of which individuals belong in the control or experimental group.

• Some hypotheses about human health are difficult to test with experiments. These hypotheses may be tested using a correlational approach, which looks for associations between two factors. A correlation can show a relationship between two factors, but it does not eliminate all alternative hypotheses.

• Statistics help scientists evaluate the results of their experiments, by determining if results appear to reflect the true effect of an experimental treatment on a sample of a population. A statistically significant result is one that is very unlikely to be due to chance differences between the experimental and control group. A statistical test indicates the role chance plays in the experimental results; this is called sampling error. Even when an experimental result is highly significant, hypotheses are tested multiple times before scientists come to consensus on the true effect of a treatment.

• Primary sources of information are experimental results published in professional journals and reviewed by other scientists before publication. Most people get their scientific information from secondary sources, such as the news media. Being able to evaluate science from these sources is an important skill. Anecdotal evidence is an unreliable means of evaluating information, and media sources are of variable quality—distinguishing between news stories and advertisements is important when evaluating the reliability of information. The Internet is a rich source of information, but users should look for clues to a particular Web site's credibility.

• Stories about science should be carefully evaluated for information on the actual study performed, the universality of the claims made by the researchers, and other studies on the same subject. Sometimes confusing stories about scientific information are a reflection of controversy within the scientific field itself.

0 0

Post a comment