Our goal in writing this book was to generate an accessible and relatively complete introduction for undergraduates to the use of statistics in the biological sciences. The text is designed for a one-quarter or one semester class in introductory statistics for the life sciences. The target audience is sophomore and junior biology environmental students, biochemistry, and health science majors. Appropriate prerequisites include some coursework in biology as well as a foundation in algebra (but not calculus). Examples are taken from many areas in the life sciences including genetics, physiology, ecology, agriculture, and medicine.
This text emphasizes the relationships among probability, probability distributions, and hypothesis testing. We have tried to highlight the expected value of various test statistics under the null and research hypotheses as a way to understand the methodology of hypothesis testing. In addition, we have incorporated nonparametric alternatives to many situations along with the standard parametric analysis. These nonparametric techniques are included because undergraduate student projects often have small sample sizes that preclude parametric analysis and because the development of the nonparametric tests is usually readily understandable for students with modest math backgrounds. The nonparametrics can be skipped or skimmed without any loss of continuity.
We have tried to include interesting and easily understandable examples with each concept. The problems at the end of each chapter have a range of difficulty and come from a variety of disciplines. Most are not real life examples but are realistic in their design and data values. The end-of-chapter problems are randomized within each chapter to require the student to choose the appropriate analysis. Many undergraduate texts present a concept or test and immediately give all the problems that can be solved by that technique. This approach prevents students from having to make the real life decision about the appropriate analysis. We believe this decision making is a critical skill in the introduction of statistical analysis and have provided a large number of opportunities to practice and develop this skill.
The material in our textbook can be supported by a wide variety of statistical packages and ancillary materials. The selection of these support materials is usually dictated by person interests and cost considerations. Here we wish to highlight four items that we have found quite useful in teaching biostatistics to undergraduates. Presently we use Statview Software developed by SAS Institute Inc. in the laboratory sessions of our course. This software is easy to use, relatively flexible and can complete nearly al the statistical techniques presented in our text. Information regarding this program can be found at www.Statview.com. A second very useful and accessible statistical package is MINITAB by Minitab Inc. Information regarding this statistical software can be found at www.minitab.com.
For student purchase we have used Texas Instrument calculators ranging from the TI-35 to the TI-83 model. The price range for those calculators in considerable and might clearly be a factor in choosing a required calculator for a particular course. Although calculators such as the TI 35 do less automatically, they sometimes give the student clearer insights into the statistical tests by requiring a few more computational steps. The ease of computation afforded by computer programs or sophisticated calculators sometimes leads to a "black box" mentality about statistics and their calculation.
Finally, for both students and instructors we recommend A Handbook of Small Data Sets, Hand, D.J. et al., editors. 1994 Chapman & Hall. London. This book contains 510 small data sets ranging from the numbers of Prussian military person killed by horse kicks from 1875-1894 (data set #283) to shape of bead wok on leather good of Shoshoni Indians (data set #150). The data sets are interesting, manageable, and amendable to statistical analysis using techniques presented in our text. While sixteen of the data sets from the Handbook were utilized as examples or problems in our text, there are many others that could serve as engaging and useful practice problems.
The material for this text derives principally from a required biostatistics course on of us (Glover) has taught to undergraduates for more than twenty years and from a second course in nonparametric statistics and field data analysis that the other of us (Mitchell) has taught more recently during several term abroad programs to Queensland, Australia. Recent shifts in undergraduate curricula have de-emphasized calculus for biology students and are now highlighting statistical analysis as a fundamental quantitative skill. Hopefully our text will make teaching and learning that skill somewhat less arduous.
Acknowledgments
We are grateful to the following reviewers of this first edition for their helpful comments and suggestions:
Tyler Haynes
Saginaw Valley State
University
William A. Hayes
Delta State University
Andrew Jay Tierman
Saginaw Valley
State University
John E. Weinstein
Texas A & M
University, Commerce
Brenda L. Young
Daemen College