Making inferences about a population of interest is often essential for scientific inquiry. This can be important both in traditional social science research, for example when using probability surveys, as well as in new forms of data, for example sampling from large amounts of data. Furthermore, understanding sampling can enable researchers to appraise the strengths and limitations of research methods and guide them towards more valid and robust ways of collecting and analysing data.
This course provides an overview of sampling techniques frequently used in survey designs. In particular, it focuses on the principles of designing and selecting samples of individuals. These principles are also discussed in terms of the effects on inference to the population of interest, the key goal of survey research. We will also briefly discuss weighting and its role in making inferences about the population.
By the end of the workshop the participants will:
- Learn about the main types of probability and non-probability sample designs
- Will learn about sampling frames
- Will understand how sampling error is related with standard errors and confidence intervals
- Will understand the concept of design effects and effective sample size
- Will understand the basics of weighting
- Introduction to probability and non-probability sampling
- Sampling frames
- Sampling error and confidence intervals
- Design effects and sample size
Who will benefit
Sampling is an essential part of research in the social sciences. This course will benefit anyone working with social data or planning to collect their own data. Knowing the basics of sampling will enable practitioners to understand the strengths and limitations of the data they use or want to collect.
- Know the main types of sampling approaches in survey research
- Understand how probability sampling can lead to inferences about populations of interest
- Understand the relationship between sampling, sampling error and design effects
- Understand the relationship between sample sizes, design effects and sampling strategies
- Understand the basics of weighting
Alexandru Cernat is a lecturer in social statistics at the University of Manchester. He received a PhD in survey methodology from the University of Essex and was a post-doc with NCRM. His expertise covers: latent variable modelling, non-response, new forms of data, longitudinal data design, longitudinal analysis. You can find more about his research and activities at www.alexcernat.com
This course contributes 6 hours to the MRS CPD programme