overview

Statistics 3163 is an intermediate course in statistics, intended as a second course for students who have completed an introductory statistics course such as STA 2023. The course quickly reviews the techniques and ideas covered in STA 2023, including descriptive statistics, tests and comparisons for means and proportions, linear regression, and chi square. The course then builds on these techniques to develop ANOVA and multiple regression modeling. The emphasis of the course is on working with (real, messy) data and understanding the complete cycle of a statistical analysis, including designing a study, collecting data, organizing data in an appropriate and useable format, using statistical software to analyze data, presenting findings orally, graphically and in written reports, and using the results of a study to inform further work. We will use case studies from economics, biology, psychology, environmental science, and other areas to emphasize this process.

This course will involve substantial computation both by hand and using software, and is expected to satisfy the Gordon Rule (computation) requirement by enhancing students computational skills via the development and use of statistical formulas based on the theory of mathematical probability and the implementation of statistical computations via software. Students must understand the underlying computational rules to make proper use of the built-in software features. Computational proficiency will be emphasized in the daily homework and computational proficiency and conceptual understanding will be emphasized in roughly equal proportions on the in-class midterm and final. Since computational ability and understanding are prerequisite to higher level conceptual understanding, it will be impossible for a student to pass the course without mastering the computational component of the class.

books

The following text is required:

Lomax, R. G. (2007).  Statistical concepts: A second course (3rd ed).  New York:  Routledge.

In addition, supplementary readings will be assigned as appropriate.

goals

This is an intermediate course in statistics, and is concerned with the learning to apply basic methods of statistical data analysis to real situations and data.  As is typical in the Honors College, we are also concerned with the wider issues of communicating effectively and developing broad critical thinking skills.  In the first category, the goals of the course are for you to (by the end of the semester):

• think critically about data;

• correctly apply and interpret standard statistical inference procedures (tests and comparisons for means and proportions, linear regression, chi-square and one- and two-way ANOVA) to both simple and complex data sets;

• have experience in producing data that give answers to reasonable questions;

• be able to build appropriate multiple linear regression models useful in a variety of situations.

In the second category, you should further develop those skills intrinsic to effective citizenship and to the study of mathematics and the sciences. 

Over the course of the semester you should:

• develop skills in communicating effectively, including writing precisely about technical things and making appropriate use of graphs and other visual displays of data;

• improve logical thinking and problem solving skills; and

• develop skills with statistical tools such as SPSS.

policies

Attendance and tardiness:  You are expected to be in class every day.  Coming to class late is disruptive and rude, so please be on time.  We don't take attendance, but there is a strong correlation between attendance and final grades.  Missing class more than once or twice during the semester is likely to affect your grade, either directly or indirectly. 

Homework:  Homework will be assigned at each class, and should be kept organized in a 3-ring notebook. We will have occasional quizzes over homework exercises.  Many homework exercises will emphasize computational proficiency and choosing appropriate statistical tools.  In addition, we will use case studies as a basis for some of our class discussion, and for these there will be more open ended homework assignments, which may be collected in whole or in part, may be discussed in class, or may be presented by students to the class.

Projects:  You will complete one to two projects over the course of the semester.  The projects will emphasize the higher order conceptual skills you are learning, with particular emphasis on the entire process of gathering, organizing, analyzing and reporting on data. You should follow the writing guidelines provided.  Reports will be graded not only on statistical correctness, but also on clarity, style and presentation. 

Late Policy: Work is due at the beginning of class on the announced due date.  We will accept late work, for reduced credit, until we have graded an assignment or project.  We generally grade materials within a couple days of collecting them, and sometimes grade them the same day they are collected.  Expect to lose approximately 10% for each day an assignment is late.

Tests:  We will have four exams, including three midterms and a comprehensive final.  The tests will emphasize both computational proficiency and statistical concepts.  Because computational understanding is necessary for an in-depth understanding of many statistical concepts, it will be difficult if not impossible to pass the midterm and final without a high level of computational skill.

Computers and Software:  We will be using SPSS during class.  You are also welcome to use Fathom, Excel, or other statistical software with which you are already familiar.  The lab in AD 122 has all the software you will need.  You may also choose to download a trial version of SPSS for use outside of class or purchase the SPSS Student-Version CD-ROM. Visit www.spss.com for the free 30 day download (or buy the book and/or CD).

Workload and Assistance:  You should expect to spend 6 to 9 hours each week, outside of class, on the course material.  This includes reading, completing homework assignments and projects, and studying for quizzes and exams.  Some weeks (those in which an exam is scheduled, for instance) may require slightly more of your time, other weeks may require slightly less, but on average, budget 6 to 9 hours each week. If at any time during the semester you find yourself having difficulty with the material, please come see us as soon as possible.  You will find it much easier to learn new topics if you consistently keep up with the course material and homework problems.

Honor Code.  You are expected to adhere to the Honor Code (see http://www.fau.edu/divdept/honcol/students/honorcode.html).  You must document your sources, whether they are human, print or web-based.  Signing your name to work to which you have not contributed, particularly group project reports, is a violation of the Honor Code.  Signing another�s name to work to which he or she has not contributed is also a violation of the Honor Code.  Failure to abide by the Honor Code will generally result in a 0 for the assignment, project, quiz or test.  A second violation will generally result in an F for the course.  We will be explicit about when you may collaborate and when you may not.  We expect the work you turn in to be yours and yours alone.

grades

Course grades will be determined as follows.

From exams, 55 percent. This includes 25% for the final, and 10% for each of three midterms.  Exams may include computations, multiple-choice, short-answer, and essay questions.  Exams will cover both readings and lecture/discussion materials.  The final exam will be cumulative.

From class participation, quizzes, and written in-class exercises, 25 percent.  Though we do not as a rule take attendance, in-class exercises will occasionally be administered for which attendance (or a signed medical excuse) will be required to earn credit.

From two class projects, 20 percent.  Class projects will be due on February 24th and April 14th. The projects will include ANOVA-based and regression-based data analysis, together with  interpretations of the results.

schedule

Please consult these pages frequently, as all dates are subject to change.  A current version of the syllabus can be found at http://wise.fau.edu/~lanning/classes/intermstats. 

Meeting	Day	Date	Topic	Readings
1	T	6-Jan	PRETEST	---
2	Th	8-Jan	Discussion of short papers and pretest	tba
3	T	13-Jan	Confidence intervals and hypothesis tests - Review	tba
4	Th	15-Jan	Chi-square - Review	tba
5	T	20-Jan	One-Way ANOVA	1.1-1.4.4
6	Th	22-Jan	Assumptions, alternatives, and applications	1.5-1.8
7	T	27-Jan	Planned comparisons	2.1-2.2.4
8	Th	29-Jan	Post-hoc comparisons / review	2.2.5-2.4
9	T	3-Feb	TEST 1	---
10	Th	5-Feb	Two-way ANOVA	3.1-3.1.10
11 12	T Th	10-Feb 12-Feb	Higher order ANOVA	3.2-3.5
13	T	17-Feb	Introduction to ANCOVA	4.1-4.4, 4.8
14	Th	19-Feb	Random effects models	5.1-5.2
15	T	24-Feb	Repeated measures	5.3-5.6, Project
16	Th	26-Feb	TEST 2	---
	T	3-Mar	Spring break	---
	Th	5-Mar	Spring break	---
17 18	T Th	10-Mar 12-Mar	Hierarchical (nested) models & randomized block models	6.1-6.7
19	T	17-Mar	Introduction to linear regression	7.1-7.4
20	Th	19-Mar	TEST 3
21	T	24-Mar	Partial correlation & Introduction to multiple regression	8.1 - 8.2.2
22	Th	26-Mar	Multiple regression continued	8.2.3-8.3
23	T	31-Mar	Other regression models	8.3-8.6
24	Th	2-Apr	Time series analysis	tba
25	T	7-Apr	Multivariate techniques: Factor analysis
26	Th	9-Apr	Multivariate techniques: Latent structure analysis	tba
27	T	14-Apr	Multivariate techniques: Multidimensional scaling	tba, Project
28 29	Th T	16-Apr 21-Apr	Review and integration	tba
--		27- Apr	Final exam (10:30 AM - 1 PM)	---

Note: In the readings column, numbers refer to chapters in Lomax (2007). tba = reading to be assigned.