The Research and Expertise Centre for Survey Methodology (RECSM) presents its "New Developments in Survey Methodology" Course Series, which will be held throughout the 2010-2011 academic year at Universitat Pompeu Fabra in Barcelona. The courses will cover some of the latest advancements in survey methodology and are addressed for intermediate and advanced students from the academic, governmental and commercial sectors.
The three planned courses for this year are the following:
Further details about the upcoming courses will be posted on the RECSM website throughout the year. Fore more information, please contact us at recsm@upf.edu.
COURSE 1: "Attitudes and their measurement using subjective reference points" (October 27-28, 2010)
Instructor: Prof. Dr. Ákos Münnich, Director of the Institute of Psychology and Head of the Department of Social and Work Psychology of the University of Debrecen, Hungary (CV).
(To view or download complete information on the course in .pdf format please click here)
COURSE 2: "Strategies in comparative analysis: multi-level, multi-group and dummy approaches" (February 9-10, 2011)
Instructor: Dr. Bart Meuleman, Assistant Professor of research methodology at the Centre for Sociological Research - University of Leuven, Belgium.
(To view or download complete information on the course in .pdf format please click here)
COURSE 3: "Probability sampling vs. non-probability sampling: What should we use and why?"
(To view or download complete information on the course in .pdf format please click here)
Instructor: Dr. Matthias Ganninger, Statistician at the GESIS - Leibniz Institute for the Social Sciences, Mannheim (Germany).
Date: May 30-31, 2011, from 10h00-13h00 and 14h00-17h30.
Place: Universitat Pompeu Fabra, Campus de la Ciutadella. Barcelona, Spain.
***This course is complemented by the seminar "Probability sampling vs. non-probability sampling", which will take place on June 1, 2011 at Universitat Pompeu Fabra. It is advised that all course participants attend the seminar as well. For more information on the seminar, please click here.***
Short description of the course:
Besides the well-known simple random sampling designs (with and without replacement) there exists a wide range of alternative selection schemes which are used in practice. Most of them have their roots in probabilistic theory, some don’t. All of them have their merits and drawbacks which must be taken into account at the planning stage as well as in the data analysis process.
There exist certain criteria which can be used to distinguish between these candidate sample designs. At the heart of these criteria lies the rationale that any sample design should be evaluated together with the estimator(s) to be used for analyses after data collection. This notion of a strategy narrows the focus of quality criteria to be applied in the evaluation to two concepts: bias and precision.
The course will focus on the interaction of diverse sample designs with commonly used estimators and useful methods to robust criteria for the selection of sampling strategies. It will reflect recent findings on the use of sample designs, weighting and variance estimation techniques. The course will help participants gain an understanding of the impact of the sample design on estimators and show pathways to achieve increased quality of their research results through practical examples.
Detailed course outline:
Day 1:
1) Introduction
Overview of the emergence and development of the (very young) field of statistics and its impact on modern society.
2) Selected basic sample designs
Introduction of the sample designs most commonly used as a basis for comparison, such as simple random sampling (with and without replacement) probability proportional to size sampling, stratified random sampling, etc.
3) Unequal probability sampling
Under certain circumstances unequal probability sampling may be applied. Some relevant sample designs will be introduced and their specific merits and shortcomings will be discussed.
4) Estimation and weighting
Basic aspects of design weighting, inferential statistical techniques, simple design-based estimation methods such as the Horvitz-Thompson or the Hansen-Hurwitz estimator will be introduced and discussed.
5) Discussion
Day 2:
1) Merits and drawbacks of multi-stage sampling
Introduction of the class of multi-stage sample designs; discussion of their merits e.g. in fieldwork and their drawbacks.
2) Non-probability sample designs
A wide range of non-probability sample designs used in practical applications is discussed with a focus on quota samples.
3) Design effects in multi-stage sampling
Specific characteristics of multi-stage sample designs include the emergence of design effects. Different approaches to estimation and a priori prediction are introduced and their use in practical applications is discussed.
4) Interviewer effects in sample surveys
Related to design effects, interviewer effects emerge in all types of surveys that incorporate an interviewer. Interactions of design and interviewer effects are delineated and discussed.
5) Discussion
About the instructor:
Dr. Matthias Ganninger received his PhD. in Statistics from the University of Trier in 2009. He currently works at the GESIS-Leibniz Institute for the Social Sciences (Mannheim, Germany), where he is Chair of the Sampling Expert Panel of the European Social Survey, among other positions. He has held positions at the German Institute for Economic Research (DIW – Berlin) and the University of Konstanz. His main research interests are in the areas of sampling design effects and census methodologies, topics in which he has given numerous courses and seminars. His work has been published in important journals such as Biometrika and Survey Methodology.
Participation:
The course is directed to anyone involved and/or interested in sampling methodology, especially experts coming from survey research commercial organizations, and students of statistics, political science and sociology. A maximum of 30 people can participate. Participants will be admitted on a first-come first-serve basis: the first 30 people who sign up and pay the registration fee will be accepted.
Registration procedure:
Please send an email to recsm@upf.edu, indicating your interest in registering for the course and including your name, institution, address, telephone and email address. As soon as this email is received, we will send you information on how to proceed with the payment. Registration will only be considered complete when payment for the course has been received by RECSM by bank transfer.
Costs:
Participants from commercial organizations: € 400
Participants working at Universities (Researchers, Professors): € 200
Students (from any University): € 80
Deadline for registration: May 25th, 2011.
Further information: If you have any questions about the course, its program or its registration procedure, please contact us at recsm@upf.edu