Survey Analysis

25-26 SURV

Engels

Description

In September 2011, the BBC News website celebrated the UK Office of National Statistics' 70-year anniversary by running a story on surveys. Online readers commenting on stories, launched a range of pertinent questions: on the selection of respondents ("How come I've never been asked?”), on the wording of questions and answers, on the variability of answers within (depending on the mood) and between people, on the truthfulness of given answers…

This course aims to make you familiar with the approach of survey methodologists to these types of questions. On the first day, we will discuss "representation errors", on the second, "measurement errors" and throughout the 2.5 days participants will work on a small project in survey analysis, which is presented on the final day.

Representation of a population can be achieved by probability sampling. The probability of including a specific unit need not be identical across the sample: stratified, clustered, and other unequal probability samples are, in fact, more common than simple random samples. Ignoring such "complex" sampling designs can seriously bias parameter estimates and their standard errors. Day one of the course therefore studies the effect of complex sampling on estimates and explains how they can be handled in applied data analysis. Nonresponse effects form a particularly tricky issue which require additional assumptions and models: the theory behind them is given as a separate topic.

Measurement concerns the extent to which obtained numbers correspond to the true values of interest. For example, when assessing drug compliance, asking `how often people visited a doctor in the past year’ may suffer from imperfect recall and the desire to give a pleasing answer, so that the answers are not perfectly reliable. On day two of the course, we will practice accounting for validity and reliability when designing survey questions. We will further study how validity and reliability of survey questions can be evaluated in practice, and consider additional complications that may arise from comparing groups such as countries or different language speakers.

In summary, these two and a half days will give participants an overview of issues arising when designing and analyzing surveys, and will offer some hands on experience running related analyses. We will be pointing towards further reading in books, software, and other resources.

Target Audience

(Post)graduates and practitioners who wish to analyze complex surveys, to design a survey, or to acquire practice with general topics in the field survey methodology.

Course prerequisites

Participants are expected to be familiar with basic data analysis methods up to the level of regression analysis, and to show personal initiative in the class exercises. They can bring their own laptop if they wish. A cursory familiarity with R is desirable (literature recommendation is given on the website).

Exam / Certificate

Participants can, if they wish, receive a grade for their project presentation and the work behind it. If you succeed in this test a certificate from Ghent University is issued. This course can be incorporated in a doctoral training program.

Type of course

This is an on campus course. We offer blended learning options if, exceptionally, you can't attend a session on campus.

Schedule

03/06/2026: from 10am until 4 pm
04/06/2026: from 10am until 4 pm
25/06/2026: from 10am until 4 pm

Venue

Faculty of Science, Campus Sterre, Krijgslaan 281, 9000 Ghent,

Course material

Access to slides

Recommended Books

Book 1: Lumley, T. (2011). Complex surveys: A guide to analysis using R. New York: Wiley.

Book 2: Saris, W. E., & Gallhofer, I. N. (2007). Design, evaluation, and analysis of questionnaires for survey research. New York: Wiley-Interscience.

Fees

The participation fee is 975 EUR for participants from the private sector. Reduced prices apply to students and staff from non-profit, social profit, and government organizations.

Industry, private sector, profession*: € 975
Non profit, government, higher education staff, (Doctoral) students, unemployed: € 495

*If two or more employees from the same company enrol simultaneously for this course a reduction of 20% on the course fee is taken into account starting from the second enrolment.

Registration

To register, add the course below to your shopping cart and proceed to checkout.

Is this your first registration for a Beta Academy course? In that case, you will need to create an account first. Afterward, you will receive a confirmation email to activate your account on the academy platform. Once activated, you can return to your shopping cart to complete your course registration. If you do not receive a confirmation email for your order, please contact our Academy for Lifelong Learning at science-academy@ugent.be.

Are you currently on the Nova-academy website? To proceed with the registration, simply click on the "More information" box located on the left side.

UGent PhD students

Doctoral School pays for your course on the condition that you sign the attendance list for each lesson. If you are absent, please notify our academy in advance by email and provide the necessary documents.

By registering for a course or event organized by the Science Academy, you agree to the cancellation procedure that you can find on our website.