Stat 100 Syllabus - Prof. Miratrix - Harvard Fall 2014 PDF

Preview

Summary

Stat 100 Syllabus - Prof. Miratrix - Harvard Fall 2014...

Description

Statistics 100 Syllabus Statistics 100 introduces the basic concepts of statistical inference, with an emphasis on data analysis and visualization instead of theory. Throughout this semester, we’ll learn how to effectively collect, understand, and analyze data. In particular we will learn how to make inferences and conclusions about real world phenomena. One part of the course will be learning how to analyze data with R, a statistical software package that is now widely used. By the end of the term, you will be able to read and critique statistical arguments. You will also have the capability to collect and analyze data on your own. This course will be co-taught with the Extension School course Stat E-100. This means lectures will be during the normal day, but professionally videotaped and posted on-line within a day. You will have access to these videos, but are expect to attend lecture. On-line help forums will be joint. Sections for 100 and E-100 will be, however, separate. We reserve the right to change details of this syllabus so we can adapt the course to better serve the overall needs of the students taking it. (But changes are unlikely to occur.) Instructor: Luke Miratrix, Assistant Professor of Statistics Email: [email protected] Intended Audience Statistics 100 is designed for those who intend to do work or have interest in the social sciences, humanities or ‘liberal arts’ such as government, anthropology, education, language, art, or literature. We do not assume that students have had a prior statistics course. If you have had AP statistics in secondary school, that is fine, but many of the concepts of course will feel familiar. However, we hope to explore those concepts more fully. Furthermore, this course is far more hands-on than a typical AP Statistics course. There is no mathematics pre-requisite beyond secondary school algebra although some familiarity with elementary algebraic notation is assumed. Graduate students who wish to tailor the course to relate to their research are invited to discuss this with the instructor. Comparing Stat 100 to 101, 102, and 104 Stat 100, Stat 101, Stat 102 and Stat 104 are the primary introduction to statistics courses offered by the Harvard Department of Statistics. Stat 101 and 102 are the most specialized, with Stat 101 is designed specifically for students intending to concentrate in psychology and Stat 102 is designed specifically for students who intend to concentrate in a life sciences discipline. Stat 104 focuses more on economics and policy analysis and Stat 100 focuses more on social sciences and the humanities. All of the courses are roughly the same level of difficulty, but the difficulty comes from different sources (e.g. differing degrees of mathematical depth, computer use, number of topics versus depth of coverage, and so forth). Both 100 and 104 offer similar exposure to basic data analysis, probability, statistical inference and building models from data, although Stat 100 will be far more applied and hands-on. Please note that Stat 104 has been revised from previous offerings and will now move at an even faster pace with a bit more mathematical exposition than before. For 1

example, calculus will be used to explain concepts. There will also be a sizable unit on basic probability theory. It also focuses on the formulas behind a wide variety of statistical tests. Stat 100 does not have these things. Therefore, Stat 100 is more suitable for those students not as comfortable with mathematical notation and equations. Stat 100 will, by contrast, spend more time on conceptual understanding and using computers for statistical work. This will include some basic programming in R, a statistical package widely in use. Objectives After finishing this course, you should have: 1. The ability to summarize and present data numerically and visually. 2. The ability to read an academic paper containing standard statistical analyses and understand the statistical claims made. 3. The knowledge of which statistical methods to use in which situations, the technological expertise to use the appropriate method(s), and the understanding necessary to interpret and present the results correctly, effectively, and in context. 4. The ability to use R, a statistical software package, to analyze real data. 5. An understanding of the importance of data collection, and some common difficulties and pitfalls with it. 6. The ability to think critically about data-based claims and quantitative arguments. 7. The ability to learn new statistical analysis techniques on your own. Textbook The text is Statistics: Unlocking the Power of Data by Lock, Lock, Lock Morgan, Lock, and Lock (cited in this course as Lock5 ). You should get the textbook along with WileyPlus, which will be used for part of the homework and some basic review. There are electronic and physical versions of the textbook available. Lectures and Learning Catalytics Each lecture will use the interactive Learning Catalytics software so as to encourage student participation and help students self-assess their progress. They will be videotaped, and you can view these recordings to help you review concepts. Learning Catalytics (LC) is software that allows students to answer questions and express opinions to lecture prompts in real-time with their smartphones or laptops. It is like “Clickers,” which are often used in other courses. LC helps students self-assess their own progress and understanding. It also allow me to adjust the presentation of the material during a lecture for increased effectiveness. We will also use LC to measure participation in class. A students will receive full credit for lecture participation if he or she is present and responds to at least 75% of the prompts. Correctness of ones answers is not part of the participation grade. 2

Sections There will be weekly discussion sections to work through examples, practice using the statistical software, and review difficult lecture material. They are not mandatory, but you are strongly encouraged to attend. They will begin the second week of the course. On-Line Forum This course uses Piazza, an on-line discussion forum designed for college courses. This forum is so students can get help with the computer portion of the course as well as general homework problems. Your participation on this forum forms part of your grade. You will receive full marks if you either ask well-formed questions or help answer them throughout the course. I.e., we will focus on evidence of constructive participation, not evidence of knowledge. Computation The course will make extensive use of the statistical software package R, which runs on both PCs and Macs. The software is free and available online through www. r-project.org. You will be taught everything you need to know how to run R. R is straightforward to learn, but is sufficiently powerful and versatile to be useful for real projects that you might carry out after this course. It is used widely in both introductory courses and in research in such fields as economics, medical research, epidemiology and political science. The first problem set contains a mini-tutorial on using the software. When you run analyses using R, you should cut and paste the relevant output into your homework. Do not hand in piles of computer printouts with your homework. This course will also rely on your having an ordinary calculator for examinations and quizzes. Problem Sets The problem sets are an important opportunity for learning, perhaps the most important, because many of the concepts in the course become clear when applied to example data sets. We encourage students to consult with classmates while working on the problems, but please read the section on Collaboration, below, first. Problem sets will be distributed on the course web page each Friday. Students will generally have a week to complete them, and they will be turned in on-line. You will generally be able to write up your home with a standard word processor. Turned in work should be neatly presented. Written answers need to be full sentences. Graded problem sets will be returned the following week. The problem sets will contain a mixture of short answer questions, questions that require some calculation, and data sets to be analyzed. The data analyses will be done in R. Late problem sets will not be accepted under any circumstances. We will, however, drop your lowest problem set from your grade. Exams All exams will be closed book. Students will be allowed to bring a double-sided 8.5x11 sheet of notes to the midterm and two such sheets to the final. These sheets have to be written or assembled by the student.

3

A hand calculator is also necessary for all exams. We cannot allow you to use a cell phone calculator or laptop during the exams, for obvious reasons. We also cannot allow sharing of calculators during exams. The midterm will be 1.5 hours and the final 3 hours. The time of the final is not known yet as it depends on Harvard’s scheduling. Harvards policy on missed exams may be found online.1 Students who miss an exam due to a religious observance are entitled to a make-up exam. For any other excused absence (medical, sports or certain club activities) the remaining exam percentages will be reweighted and a make-up will not be offered. Course project An end-of-semester project is required of all 100 students. The projects are intended to be carried out in teams of two or three students, although individual projects are also acceptable. The project will consist of preparing a poster on anything statistical which interests you, and then presenting the poster to other students and Stat 100 staff during reading period. The main goal of the project is to convince your TF and instructor that you are able to apply knowledge learned through Stat 100. Some ideas: Analysis of an application of statistics to your field (or a hobby), description and application of a statistical concept that we did not discuss in class, or a statistical analysis of a data set. Details of the poster project will be forthcoming. You should plan to discuss ideas of a project with your TF or the instructor. You will be asked to submit a formal proposal in November. Grading The course letter grade will be based on the following: Component % grade When Problem sets 30% Weekly (lowest score dropped) Lecture Participation 5% Daily (w/ Learning Catalytics) On-line forum participation 5% Midterm 15% 17 October 2014 Final Project 15% 8 Dec 2014, 9:30AM-12PM Final Exam 25% Exam Period Best Show 5% Best of Final, Midterm, or Final Project Total 100% Collaboration Appropriate and judicious collaboration is encouraged. Discussion and the exchange of ideas are essential to doing academic work. For assignments in this course, you are encouraged to consult with your classmates as you work on problem sets. However, after discussions with peers, make sure that you can work through the problem yourself and ensure that any answers you submit for evaluation are the result of your own efforts. In addition, you must cite any books, articles, websites, lectures, etc that have helped you with your work using appropriate citation practices. Similarly, you must list the names of students with whom you have collaborated on problem sets. Furthermore, for any writing 1

http://static.fas.harvard.edu/registrar/ugrad_handbook/current/chapter2/attendance_ absences_etc.html.

4

assignments, if you received any help with your writing (feedback on drafts, etc), you must also acknowledge this assistance. Regrading Clerical errors will be corrected without any hassle. Other regrade requests must be submitted in writing within a week of the item’s return. To discourage “grade grubbing,” the entire item will be subject to regrading (even if the regrade request is not honored). Course Help If you need help with the course material, you have several options: 1. Go to the on-line discussion forum, Piazza, and ask for help! 2. You can always come see your TF or instructor during our office hours (or by making an appointment). 3. For quick questions or clarifications, you can send an e-mail message to your TF or instructor. Emails should generally be replied to within one full business day. 4. You can attend Study networks, where undergraduate “facilitators” who have previously taken Statistics 100 will be available to meet with students during their weekly meeting hours. More information about study networks will be provided in class. 5. If you feel like are losing touch with the course material and you need extra individual mentoring time beyond course office hours, you may want to consider contacting the Bureau of Study Counsel.2 They employ former Statistics 100 students who can serve as private tutors at a nominal hourly fee. Accommodations Students needing academic adjustments or accommodations because of a documented disability must present their Faculty Letter from the Accessible Education Office (AEO) and speak with the professor by the end of the second week of the term, September 14th. Failure to do so may result in the Course Head’s inability to respond in a timely manner. All discussions will remain confidential, although Faculty are invited to contact AEO to discuss appropriate implementation. Academic Integrity Harvard College is a community dedicated to scholarship, leadership, and service and to the principles of honesty, fairness, respect, and accountability. Citizens of this community commit to reflect upon and uphold these principles in all academic and nonacademic endeavors, and to protect and promote a culture of integrity. Cheating on exams and quizzes, plagiarism and copying others’ work on homework assignments and projects, lying about an illness or absence and other forms of academic dishonesty are a breach of trust with classmates and faculty, and will not be tolerated. See the integrity website for more.3 2 3

http://bsc.harvard.edu http://academicintegrity.fas.harvard.edu/icb/icb.do

5

Graduate Students Graduate students are expected to do a larger course project, ideally in connection with their research, and they are not required to sit for the final exam. They are expected to do the homework assignments and midterm. They are invited to make an appointment with me to talk about how to structure the course to suit their individual research needs. Acknowledgements This course and much of the material in it comes in large part from several others who generously aided me with their time and materials. In particular (alphabetically) I thank Mark E. Glickman, Dave Harrington, Kari Lock Morgan, and Michael Parzen.

6...