EDLD 613: Advanced Quantitative Research Methods Fall 2009, MRH 133, Tuesdays 5:30 – 8:20 Professor: Fred Galloway, firstname.lastname@example.org Office Phone/Fax: #275H, 619-260-7435/619-849-8175 Office Hours: Mondays and Tuesdays: 2:30 – 5:00 and by appointment
The purpose of this course is to help students develop the research skills necessary to do high quality quantitative research for their dissertations. As such, this course requires students to have successfully completed the basic doctoral-level quantitative research methods course, EDLD 607. Any students not meeting this prerequisite will need the special approval of the instructor before formally enrolling in this course. Although there are many objectives for students in this course, students will be expected to master the following four basic objectives: • • • •
learn how to formulate research questions and testable hypotheses; understand what it means to make a robust statistical inference; master the fundamentals of multiple regression analysis; and design, analyze, and present an original empirical project.
To meet these objectives, all students will be expected to attend class regularly, do the assigned readings before class, complete all assigned homework problems, pass the inclass midterm, complete and present an original empirical project, and actively contribute to class discussions. Please remember that in terms of class participation, the quality of the contribution is infinitely more important than the quantity of the contribution. In other words, monopolizing class discussions provides little, if any value-added to the class and as such, is to be avoided. Although this may seem like a large set of skills to acquire in one course, the framework of multiple regression analysis will provide us with the scaffolding necessary to do highquality quantitative research. And since this framework is used extensively in economics, political science, and sociology, any empirical research produced using these techniques will be easily understood by other social scientists. For those who think that some sort of shortcut exists using other research techniques, I offer you the overwhelming body of poorly conducted, non-generalizable quantitative research done by many of those who empirically study either leadership or education. In fact, the proliferation of such work has been almost solely responsible for both fields being somewhat marginalized by those with strong analytic backgrounds. To overcome this structural problem, our motto for this course will be “while it doesn’t have to be fancy or super-sophisticated, it does have to be methodologically correct.” Since there are a number of excellent texts that cover regression analysis, there is no single required book for the course; most students will already have either Statistics for Business and Economics by Anderson, Sweeney, and Williams (ASW) or some other
professor-approved, equivalent text. In addition to these readings, several papers and chapters from Econometrics, Basic and Applied, by Johnson, Johnson, and Buse (JJB) will also be shared with students. To complete the empirical project, students will make use of the popular statistical software, SPSS. This extremely popular piece of statistical software is available in at least four computer labs on campus – Serra 185, Maher 114, and Hill Hall 214 and Hill Hall 216 – so there is no need for students to purchase the software themselves. Although SPSS is supported in the first two computer labs, students are expected to master the basics of SPSS by working through the tutorials provided and helping each other out with any problems that may arise. Of course, the instructor will also be available to help, but since the software uses simple pull-down menus, students are encouraged to solve their own software problems whenever possible. To assess student performance in the course, there will be several methods of evaluation. As mentioned earlier, students will be expected to pass the midterm exam, complete and present an original empirical project, and actively participate in class. To help students stay current with the material, problems will be regularly assigned during class and students are strongly urged to complete them before the midterm. However, given the experience most students had last semester with Aplia, problems in the book will be used instead of the problems associated with Aplia. Although these pre-midterm problems will not be handed in and graded, they will prepare students for the exam and will be discussed in detail during the review session. The relative weights for these evaluative components are listed below: • • •
Midterm Empirical Project Project Presentation and Class Participation
40% 40% 20%
To help students stay on top of the course material, on the following page is a weekly schedule of readings and lecture coverage. Although absences are to be expected, the nature of this particular course puts a real premium on attendance, so students are strongly encouraged to attend whenever possible. If for some reason, you must be absent during the midterm, if at all possible please make arrangements with the instructor before the scheduled date of the midterm.
Schedule of Readings, Lecture Coverage, and Course Happenings Date
Lecture Coverage Course structure and overview
Lumina Paper 1 causality, intro
to advanced quantitative methods Minimization of least squares and linear regression theory Correlation techniques
Multiple regression theory
Tests of linear restrictions
Test of linear restrictions
Functional form, specification error
Research design and
November 8 2
MIDTERM REVIEW (10 -12)
NO CLASS SCHEDULED
Data problems: measurement error and missing observations Empirical project troubleshooting
EMPIRICAL PROJECTS DUE
This paper can be found at the following website: www.luminafoundation.org/research/researchersgalloway.pdf 2 This special class, which will last from 10:00 â€“ 12:00 at a located TBD, will serve as a review session for the midterm exam the following Tuesday.