My objective is to provide both an understanding of, and hands-on experience with basic, data-centric statistics. This class is similar to a regular statistics class except that we focus the application of statistics on examples of actual studies from a wide array of biological interests: from biomedical studies and pharmacology, to ecology, genetics, physiology and related socio-economic questions.
By the end of this course, you should be able to analyze and present data, design observational and experimental studies, use probabilities to model and predict random events, and use inference procedures to test hypotheses and estimate population parameters to reach conclusions in context. What you learn in this class should help you understand broadly the methodology, results, and issues of studies presented in your other classes or in news stories. I also hope that you will come to appreciate statistics as a cool and really interesting subject.
Note that STATS 8 satisfies the General Education requirement for Category Va, Quantitative Literacy, with the following learning outcome objectives: Students should be able to
1) Identify appropriate tools for quantitative analysis of processes or events.
2) Have a basic familiarity with fundamental principles underlying quantitative descriptions of natural or social processes.
3) Be able to do one or more of the following: evaluate studies and reports that assess risk and probability in everyday life; use models of natural phenomena to make quantitative predictions of future behavior or events; use models of economic and social structures to make quantitative predictions of future behavior or events.
1. Descriptive statistics: data organization, graphs, numerical summaries, interpretation in context
2. Association: correlation, regression, two-way tables, association versus causation
3. Data collection: random samples, observational designs, experimental designs
4. Probability concepts: fundamental rules, conditional probabilities, independence
5. Probability distributions: continuous distributions, Normal distributions, sampling distributions
6. Confidence interval for a population mean: one sample and matched-pairs
7. Hypothesis test for a population mean: one sample and matched-pairs
8. Inference for several means: two-sample t interval, two-sample t test, analysis of variance
9. Inference for categorical data: chi-square test for two-way tables, chi-square test for goodness of fit
"Lectures" are split in two components for each topic covered: core content and applications.
The core content is delivered as a set of short interactive videos hosted on Canvas, to allow students to learn the concepts at their own pace. The videos need to be watched and the associated Canvas quiz taken before the "applications" part can occur (deadlines before each lecture, as shown on the quizzes). This means that basic study time should be done before coming to lecture, rather than after.
Applications take the form of active class work during the officially scheduled lecture time, to provide live guidance and feedback. Students are encouraged to work real problems out loud with their neighbors. Answers to specific questions must be submitted on the spot with the iClicker/Reef response system (part of the participation score). For topics requiring computational answers, you should bring either a graphing calculator, calculator emulator, or a device capable of accessing statistical software.
Discussion sections provide additional opportunity to practice exercises and ask questions. Real data problems (posted as worksheets within each topic page) will be solved in groups, under the guidance of the TA. Answers to specific questions must be submitted on the spot with the iClicker/Reef response system (part of the participation score).
All computations for this class are done using technology (statistical software, graphing calculator).
*** home ***