Free Trial

Safari Books Online is a digital library providing on-demand subscription access to thousands of learning resources.

Share this Page URL

Appendix - Pg. 367

Appendix A.1 Statistical Hypothesis Testing In biomedical research, we are nearly always concerned with finding some new factor that can benefit the health of the population in some way. The population in this sense could be everybody in the country or just patients with a subtype of a disease. We can- not use an entire population for our research, so we take samples where we generate or collect data from a finite number of eligible people. We always hope that our sam- ple represents the entire population for whatever variable we may be interested in. Pa- rameters that we measure in our study can be of many different types, such as the mean or the median. When we make measurements on a change in a population (e.g., how patients re- spond after a new drug therapy), whether it be using one, two, or more samples, we need to be sure that the change we see is truly due to the actual variable(s) we assume are causing the effect. Alternatively, the change may simply be due to chance.