Tel. 919-515-3067
Cell: 919-616-2273
Fax: 919-515-7333
E-mail: David_Garson@ncsu.edu
Office location: Winston 022
Mailbox location: 212 Caldwell
Office hours: Call for appt., usually prior to or following class.

Fall, 2009
PA 766 Section 001
(Temporarily PA 798)
Wednesdays, 3:00 - 5:45 pm
Location Winston 213

Schedule of due dates:
Midterm: 3:00 - 5:00 pm, Wednesday, October 7
Final project: 5 pm. Friday, Dec. 4.
Final exam: 3:00 - 5:00 pm, Wednesday, December 9

Schedule of topics and reading assignments: Listed separately on the "Weekly Topics" pages for PA 766.

PA 766 is offered in fall semesters following PA 765 in spring semesters, for three graduate credits each. PA 766 is an advanced course in research methodology as applied to public administation, with a focus on statistical modeling of public policy and management topics using specialized regression procedures, generalized linear models, hierarchical linear models, event history analysis, and structural equation modeling.

Learning objectives for the course:

1. The student will be able to explain the nature and purposes of ordinal regression, its applicability as illustrated in research problems in public administration or the student's field, its assumptions, common pitfalls in its use, related measures of significance and effect size, and diagnostic procedures, and will be able to interpret computer output for this procedure.

2. The student will be able to explain the nature and purposes of nonlinear regression, its applicability as illustrated in research problems in public administration or the student's field, its assumptions, common pitfalls in its use, related measures of significance and effect size, and diagnostic procedures, and will be able to interpret computer output for this procedure.

3. The student will be able to explain the nature and purposes of generalized linear models, its applicability as illustrated in research problems in public administration or the student's field, its assumptions, common pitfalls in its use, related measures of significance and effect size, and diagnostic procedures, and will be able to interpret computer output for this procedure.

4. The student will be able to explain the nature and purposes of event history and Cox regression models, its applicability as illustrated in research problems in public administration or the student's field, its assumptions, common pitfalls in its use, related measures of significance and effect size, and diagnostic procedures, and will be able to interpret computer output for this procedure.

5. The student will be able to explain the nature and purposes of random effects and repeated measures linear mixed models, its applicability as illustrated in research problems in public administration or the student's field, its assumptions, common pitfalls in its use, related measures of significance and effect size, and diagnostic procedures, and will be able to interpret computer output for this procedure.

6. The student will be able to explain the nature and purposes of multilevel linear mixed models in SPSS, its applicability as illustrated in research problems in public administration or the student's field, its assumptions, common pitfalls in its use, related measures of significance and effect size, and diagnostic procedures, and will be able to interpret computer output for this procedure.

7. The student will be able to explain the nature and purposes of hierarchical linear mixed modeling in HLM, its applicability as illustrated in research problems in public administration or the student's field, its assumptions, common pitfalls in its use, related measures of significance and effect size, and diagnostic procedures, and will be able to interpret computer output for this procedure.

8. The student will be able to explain the nature and purposes of confirmatory, structural, and multigroup models in structural equation modeling, their applicability as illustrated in research problems in public administration or the student's field, salient assumptions, common pitfalls, related measures of significance and effect size, and diagnostic procedures, and will be able to interpret computer output for this procedure.

Click here for the syllabus specific to the prior course, PA 765. To enroll in PA 766 students are expected to have had PA 765 or equivalent.

Note: The weekly reading, homework, and test assignments are found by clicking on "PA 766 Topics" at the class website. Note that in addition to the print readings listed on this page, you should click on links to online readings and to instructions related to textbook assignments found on the "Topics" pages.

Other related graduate courses by the instructor include PA 541 (Geographic Information Systems) and PA 803 (Advanced Research Methods).

Prerequisites:

PA 765 requires having taken ST 507 or equivalent and PA 515 or equivalent. Equivalent for ST 507 may include ST 508, ST 513, or ST 514. Additional relevant courses include ST 708 (Applied Least Squares), ST 730 (Applied Time Series), and ST 731 (Applied Multivariate Analysis). Sociology 707 (Quantitative Sociological Analysis) is suitable as a prerequisite or corequisite. Sociology 708 (Advanced Sociological Analysis) is suitable as a corequisite. PA 766 requires PA 765 or equivalent.

Keeping Current. While the instructor has tried to be helpful by placing as much as possible online at the class website, http://faculty.chass.ncsu.edu/garson/pa765/, the instructor will be updating these materials as the course progresses, particularly StatNotes. Likewise, if you printed this syllabus prior to the first day of the course, it is not official: go back and reprint the syllabus, which might also have updates.

Textbooks

There are no laboratory, transportation, or other extra expenses associated with PA 766 apart from textbooks. The course textbooks (not including the instructor's free online textbook, Statnotes, nor online materials found on the "Topics" pages) are:

• Marija J. Norušis (2009). SPSS 17.0 Statistical Procedures Companion (2009). ISBN-13: 978-0321621412(paper). Prentice Hall. $62.80 Amazon price March 2009. You will need the CD-ROM data disk which comes with this book, so be careful if you buy a used copy!

• Marija J. Norušis (2009). SPSS 17.0 Advanced Statistical Procedures Companion (2009). ISBN-13: 978-0321621429 (paper). Prentice-Hall. $55.42 Amazon price March, 2009. You will need the CD-ROM data disk which comes with this book, so be careful if you buy a used copy!

• Randall E. Schumacker and Richard G. Lomax (2004). A Beginner's Guide to Structural Equation Modeling Second Revised Edition.. Lawrence Erlbaum. ISBN 978-0805840186. $58.50 Amazon price March 2009. (Apparently the third edition promised for May 2009 has been delayed to at least 2010, according to the publisher).

• It is also assumed that you have retained the statistics and methods textbooks from prerequisite courses and that they are available for reference purposes. If you do not have such a book, you may want to own previous years' textbooks from this course (no longer assigned):
  • Barbara Byrne, Structural Equation Modeling with AMOS. Lawrence Erlbaum Associates, 2001. ISBN 0-8058-4104-0.
  • Barbara G. Tabachnick and Linda S. Fidell (2001). Using Multivariate Statistics, Fourth Edition. Needham Heights, MA: Allyn & Bacon. ISBN 0-321-05677-9. Hardcover. Datasets accompanying this textbook are free at http://www.abacon.com/tabachnick .
  • Graeme Hutcheson and Nick Sofroniou, The Multivariate Social Scientist: Introductory Statistics Using Generalized Linear Models. Thousand Oaks, CA: Sage Publications, 1996. ISBN 0 7619 5201 2 (pbk.).
  • Jacques Tacq, Multivariate Analysis Techniques in Social Science Research. Thousand Oaks, CA: Sage Publications, 1996. ISBN 076195273X (pbk.)
  • Rex B. Kline, Principles and Practice of Structural Equation Modeling. NY: Guilford Press, 1998. ISBN 1-57230-337-9 (pbk.).

• Warning! Check the class website regularly. The website is being continually updated. You may wish to reprint "StatNotes" sections on the day of the class to assure the latest version. Major updates are announced on the course listserv.

Software

Software is in the Social Science Lab. Unlike past years, the license does not cover installation on your home machine unless you are receiving paychecks from NCSU, such as for being a TA. If you are a part- or full-time NCSU employee, contact Justin_Daves@ncsu.edu about getting SPSS installed. He is also the one to contact about possible changes in the SPSS license. SPSS and AMOS will be available in all CHASS computer labs (Laundry, Tompkins, Winston, and Withers), as well as on VCL (at vcl.ncsu.edu) for remote access. Note that VCL has two SPSS images: an "All Modules" image with only two copies and a general SPSS image with two dozen copies. Please use the former only when essential (ex., for Missing Values Analysis). The walk-in labs are typically open until 9pm or 10pm Monday-Thursday, 5pm on Fridays. The Laundry lab is open on weekends as well. Note that to use VCL you must enter your NCSU id and password.

If you wish to purchase your own copy of the software, there are several student versions available: SPSS Graduate Pack $200 for 4 years; SPSS Graduate Pack $140 for 12 months; or SPSS Graduate Pack $80 for 6 months; (As of late 2007; check for actual pricing). There is also Student version and a Career Starter version that are both geared more towards undergraduates and will not do all that is required in this course, but are cheaper, For a breakdown of the different student versions see http://www.spss.com/vertical_markets/education/SPSS_student_versions.htm.

• SPSS modules needed. In this course, in addition to SPSS Base, you will need SPSS Advanced Models (for GLM, as well as survival analysis and GENLIN), SPSS Regression Models (for binary and multinomial logistic, as well as probit), Missing Values, and AMOS (for SEM). In addition, depending on your dissertation methodology, you may need SPSS Trends (for ARIMA and time series), SPSS Categories (for additional forms of scaling, nonlinear canonical correlation, correspondence analysis, or perceptual mapping), SPSS Text Analysis (for open-ended survey items), and Neural Networks (for neural network analysis).

Other software. We will be using HLM student version for the hierarchical linear modeling sections. This free software should be downloaded by the student and installed on the student's machine as explained in to "Topics" page on HLM. It cannot be guaranteed that this software will also be available on VCL or in the CHASS labs. It is recommended that the student acquire and install this software early in the semester, well prior to the scheduled HLM topics.

Homework Assignments

Textbook problems. For most weeks, there are hands-on assignments related to the textbooks, as specified on the "Topics" pages. Students should do the readings and attempt the textbook assignments before the scheduled class. Assignments are due at the class after the one in which they are discussed. Note that this means that in the first week students should do the readings and attempt the assignments for Weeks 1 and 2. Keeping "ahead" in this manner will pay off at the end of the course, leaving adequate time for finals. Note: it is not intended that you turn in answers to readings questions posed on some "Topics" pages; rather you may be called upon to answer these in class, so it is up to you whether to write out answers. If you are going to be absent on a particular day, then do turn in written answers to show you have done the work.

Printing output: To export to Word and print, highlight Output at the top of the outline pane on the left, then right-click and select Export. Export as type HTML file (the instructor found on his setup that this worked better than using type Word/RTF). After saving, load the file into Word, select View, Print Layout; then Insert, Page Numbers, add your name, and print to bring to class.

Midterm and final exams: As scheduled at top of syllabus.

Final project: A final project paper related to public or non-profit administration or policy (or other specialization for non-PA class members) is due on the last day of class. It should be quantitative in nature and its most-used statistical procedure(s) must be drawn from those covered in PA 766.

Grading The grading components are 10% text homework and participation, 30% midterm, 30% final exam, and 30% term project. No absences or incompletes are accepted for this course other than as per NCSU Attendance Regulation located at http://www.ncsu.edu/policies/academic_affairs/courses_undergrad/REG02.20.3.php. Attendance is not graded. Makeup work, if any, must be arranged within two weeks of due date at the option of the instructor, prior to two weeks before the end of classes.

Course Evaluations Course evaluations are now done online at https://classeval.ncsu.edu/. Help on course evaluations is by email at classeval@ncsu.edu..

Academic Integrity Statement

Please note the existence of the University policy on academic integrity found in the Code of Student Conduct located at http://www.ncsu.edu/policies/student_services/student_discipline/POL11.35.1.php. Academic misconduct will not be tolerated in this class. Academic misconduct may be defined as "any activity which tends to compromise the academic integrity of the institution, or subvert the educational process". I expect complete honesty in the completion of tests and assignments. It is my understanding that the student's signature on any test or assignment means that the student neither gave nor received unauthorized aid. Beyond cheating on quizzes or exams, academic misconduct also includes the submission of plagiarized work for an academic requirement. Plagiarism is "the representation of another's works or ideas as one's own; it includes the unacknowledged word for word use and/or paraphrasing of another person's work, and/or the inappropriate unacknowledged use of another person's ideas". Thus each argument made in the research papers which is not completely the student's own, must be footnoted or otherwise referenced to indicate its source.

"Plagiarism, or presenting another's works or ideas as one's own, is a form of stealing. The instructor reserves the right to examine any source used by the student before giving a grade on a paper, and to give and "incomplete" in the course if necessary, to allow time to obtain sources. Students should be prepared to show source material to the instructor for the purpose of verifying information. Academic dishonesty will not be tolerated.

Academic dishonesty includes the following offenses:

1. Claiming as your own work a paper written by another student.
2. Turning in a paper that contains paraphrases of someone else's ideas but does not give proper credit to that person for those ideas.
3. Turning in a paper that is largely a restatement in your own words of a paper written by someone else, even if you give credit to that person for those ideas. The thesis and organizing principles of a paper must be your own.
4. Turning in a paper that uses the exact words of another author without using quotation marks, even if proper credit is given in a citation, or that changes the words only slightly and claims them to be paraphrases.
5. Turning in the same paper, even in a different version, for two different courses without the permission of both professors involved.
6. Using any external source (notes, books, other students, etc.) for assistance during an in-class exam, unless given permission to do so by the professor." ---- Source: Kendra Stewart et al., Columbia College

It is the understanding and expectation of the faculty that the student's signature on any test or assignment means that the student neither gave nor received unauthorized aid.

The public administration faculty has agreed that violations of academic integrity must have consequences. Consequently, students who cheat (behaviors cited in point 1 and 6 or similar behavior) may receive at least an F in the course; other forms of dishonesty, similar to those covered in points 2 through 5 may result in at least a 0 for the assignment.

Statement Regarding Students with Disabilities

Every attempt is made in this course to conform to university policy on disabilities, as described at http://www.ncsu.edu/policies/academic_affairs/courses_undergrad/REG02.20.1.php. If you have disability-related needs, please contact the NCSU Disabilties Office at http://www.ncsu.edu/dso/ and then inform the instructor immediately.

Career Services.

Exploring career options with your major, making decisions about your major or minor, building resumes and cover letters, preparing for interviews, developing job search strategies, maximizing career fairs, and more. Make an appointment with your career contact: Sara Concini (A-H) or Woody Catoe (I-Z). Call 515-2396. University Career Center, 2100 Pullen Hall. www.ncsu.edu/career. While primarily serving undergraduates, they are happy to make individual appointments with graduate students.