Scientific Coordination
Sebastian E. Wenz
Tel: +49 221 47694-159
Tel: +49 221 47694-159
Administrative Coordination
Jacqueline Schüller
Tel: +49 0221 47694-160
Tel: +49 0221 47694-160
Please wait...
Course 5: Survey Sampling and Weighting
About
Location:
Cologne/Unter Sachsenhausen 6-8
Cologne/Unter Sachsenhausen 6-8
Course Duration
Mo: 10:00-17:00 CEST
Tu-Fr: 09:00-16:00 CEST
General Topics:
Course Level:
Format:
Software used:
Duration:
Language:
Fees:
Students: 550 €
Academics: 825 €
Commercial: 1650 €
Additional links
Lecturer(s): Simon Kühne
Course description
This practical course will cover the process of probability sampling and weighting for (social science) survey data. This includes a) sampling techniques, b) methods of weighting survey data including design weighting, nonresponse weighting, and adjustment weighting, and c) methods of parameter estimation for complex sample survey data. Please note that the course will only provide a short excursus to non-probability sampling. The emphasis of the course is more applied than theoretical, but you are expected to be comfortable with statistics and to have some experience with survey data analysis. For each topic, you will do exercises in Stata or R (based on your preference) that apply the techniques learned in the lectures. Prior knowledge in how to use Stata or R for survey data analysis is needed.
The full syllabus of the course including the day-to-day schedule will be published here in April.
Target group
You will find the course useful if:
- you have experience conducting surveys and/or analyzing survey data but have no experience with survey sampling and weighting,
- you are planning your own survey data collection and need to sample and/or weight the data.
Learning objectives
By the end of the course, you will:
- know about the most commonly used sample designs including stratified sampling, cluster sampling, and multi-stage sampling,
- know how to create design weights, nonresponse weights, and apply adjustment weighting techniques,
- know how the sample design can affect data analysis and how to incorporate complex survey designs and survey weights into parameter estimation.
Organizational structure of the course
A typical course day will consist of two thematic clusters. Each cluster starts with a lecture of around 1 hour. After that lecture, you split into small groups and work on practical exercises using Stata or R. Simon will provide and discuss a solution for each exercise in-depth afterwards. During the last half of the last course day, Simon provides guidance for your own current sampling and weighting projects.
Prerequisites
- Introductory course in statistics. No prior knowledge of sampling theory is assumed.
- Prior knowledge in Stata or R is required for this course.
- Basic understanding in survey methodology and how to handle survey data.
Software and hardware requirements
You will need to bring a laptop computer to successfully participate in this course.
You should have the following software installed on your computer before the course starts:
Stata (at least 17) packages: gsample, unique, distinct, survwgt, fre, coefplot, moremata, estout
OR
R (at least 4.3.2) packages: tidyverse, ggplot2, dplyr, haven, sampling, srvyr, survey, stratification, SamplingStrata
GESIS will provide you with short term licenses for Stata for the duration of the course if needed.
