Survey Design and Analysis
Course CEE 6623: Survey Design and Analysis. 3 units, letter-graded.
Description: Description of types of surveys commonly used in transportation demand analysis, including travel and activity diaries, computer-administered, panel, attitudinal, and stated-response surveys. Use of GPS to collect travel data. Discussion of sampling, response, experimental and quasi-experimental design, survey design, and ethical issues. Analysis methods, focusing on factor analysis, but including an overview of cluster analysis.
Goals: To develop an understanding of the types of surveys commonly used in transportation demand modeling and travel behavior research, including issues of design, sampling, and analysis. To gain hands-on experience in designing, evaluating, and analyzing such surveys.
Prerequisites: Linear Algebra and Intro to Probability and Statistics are expected; Transportation Planning (Regional Travel Demand Forecasting) is desirable.
Textbook: Recommended: Stopher, Peter (2012) Collecting, Managing, and Assessing Data Using Sample Surveys. Cambridge, UK: Cambridge University Press. Thorough compendium by a giant in the transportation field.
Dillman, Don A., Jolene D. Smyth, and Leah Melani Christian (2009) Internet, Mail, and Mixed-Mode Surveys: The Tailored Design Method, 3rd ed. Hoboken, New Jersey: John Wiley and Sons.
Also nice: Sommer, Barbara and Robert Sommer (1997) A Practical Guide to Behavioral Research: Tools and Techniques, 4th edition. New York: Oxford University Press. An excellent and not too expensive book by UC Davis faculty, covering all aspects from ethics, library searches, observation, interview, questionnaire design and administration, analysis, and presentation of results.
Richardson, A. J., E. S. Ampt, and A. H. Meyburg (1995) Survey Methods for Transport Planning. Melbourne, Australia: Eucalyptus Press. Available, by permission of the first author, on the class web site.
Additional references to be provided.
Instructor: Patricia Mokhtarian, Professor
School of Civil and Environmental Engineering
322 SEB, (404) 385-1443, firstname.lastname@example.org
Grading: Grades will be based on three substantial homework assignments and a final exam; each will count 25% of the final grade. In addition, confirmation of satisfactory completion of the first two modules of the online “research with human subjects” course is required to receive a final grade. Assignments will include critiques of existing survey and study designs, design of your own survey, and computer analysis of survey data using methods taught in this course.
Teaming: Teaming with one other person is allowed on the HW, at your choice. That is, you may team or not team, you choose your teammate (if any), and you are free to change the arrangement from one assignment to the next. Teamed assignments will receive a single grade for the team, and will be graded to the same standards as un-teamed assignments. Each member of the team is expected to engage thoroughly in, and to make substantive contributions to, all aspects of the assignment.
My general policy is not to allow three-person teams, because that dilutes the workload, and hence the understanding of the material and therefore the pedagogical value of the assignment, too much. This sometimes has the unfortunate result that someone who wants to team is the “odd person out”, when everyone else is either already paired off or does not wish to team. In such cases, someone must end up unhappy, and for both pedagogical and social-psychological reasons I would rather force someone not to be on a team who wants to be on one, than to force someone to be on a team who doesn’t want to be. In my philosophy, teaming is a “bonus”, not an automatic right. So… if you want to team, start forming the team early.
Deadlines: In the past I’ve been relatively relaxed about deadlines for this class, but I’ve noticed a considerable increase in the slippage over the years. So I am currently trying a graduated late-penalty system that works as follows:
1. The HW will be distributed with a posted official due date. Unless otherwise specified, HW will be due before the start of class on the day in question. In terms of keeping up with the class, it will be in your best interests to meet that due date. The next assignment will generally be distributed on or before the date that the previous assignment is officially due.
2. I will generally flex that due date by one or two lecture periods (according to the table below), without penalty.
3. Beyond that point, the basic principle is that I will penalize your grade by “a third of a letter” for each partial or complete time period that you are late, where a time period is generally defined as the class start time on one regularly-scheduled class day to the class start time on the next regularly-scheduled class day. However, in any given term there may be variations in view of holidays, my absences or other considerations. The penalty kicks in even if the HW is turned in later that same day, but after the class start time (or the otherwise-indicated deadline).
I realize (to my dismay) that Georgia Tech does not allow “+”s and “-“s to be given, only straight letter grades. However, I am still able to give interim grades and penalties in those finer-grained increments; I will simply have to round off at the end. So, for example, if you would have made A (4.0)s on all assignments and the exam, but turn in HW1 and HW2 as much as two weeks late, you will receive “B+”s for those assignments but will still have an A for the course: the average of 3.3, 3.3, 4.0 and 4.0 is 3.65 which, using the formula Y = 10 (X+5.5) (where X is the score on a 4-point scale and Y is the score on a 100-point scale), translates to a 91.5. On the other hand, if you would only have made “A-” (3.7)s on all assignments and turn in HWs 1 & 2 two weeks late (thereby receiving Bs on them), your final average would be 3.35, which translates to 88.5, which would be a B.
Policy on redo's, variations, or extra credit assignments: None allowed
I spend a great deal of time grading each HW, generally writing numerous comments indicating problems, missing items, etc. This is of course intended to be an additional learning experience, as well as providing you with a clear understanding of the reason for the grade you earned. However, you can imagine that by the time you've read my comments, I've practically given you the answers I'm looking for, and therefore it would not be appropriate to offer you the chance to re-do the assignment if you are not satisfied with your grade. For the same reason, it is considered cheating for you to read my comments on anyone's previously-graded assignment before turning in your own. As I also hope you can imagine, it takes considerable effort to set up each HW assignment, develop the expected answers myself, prepare the grading template (checklist of things I look for), etc. This is why I give basically the same assignments each year (with minor modifications), and why it is especially wrong for you to review someone's previously-graded assignment in advance. It is also why I am not willing to offer make-up assignments in the event that you are dissatisfied with your grade. Thus, the bottom line here is that you are only going to get the assignments I mention at the beginning of the term – the same ones everyone else gets – and you are only going to get one shot at each of them.
Topics (approx. number of 1½-hour lectures):
1. Course overview, the scientific method, and intro. to types of surveys (2)
2. Travel diaries (origin - destination surveys) and activity diaries (2)
3. Survey administration methods (1)
4. Attitudinal survey design (6)
5. Response rates, errors, sampling, non-response (5)
6. Factor analysis (4)
7. Cluster analysis (1)
8. Quasi-experimental design principles (2)
9. Systematic biases in everyday judgment (1)
10. Ethics in survey research (1)
11. Panel surveys (1)
12. Stated-response surveys and experimental design (1)
13. Qualitative methods (1)
14. Guest lectures (1-4)
Computer Usage: Analysis of data sets using multivariate statistical software (such as SPSS, R, SAS)