Skip to content

Sampling: How to make decisions on selections in development evaluation

Part of
Online Sessions
September 25 - 29
Recommended for
Evaluators, Practitioners


In almost all evaluations, it is necessary to make decisions about selections: which documents should be analyzed, which of the stakeholder groups should be included in the evaluation process, from whom is it possible to collect needed information, who should be invited to the closing session etc.  This workshop is on the several different challenges and methods to make such kind of decisions for drawing samples from any kind of populations. This includes both non-probability sampling by using different kind of criteria as well as probability sampling using inference statistics to produce representativeness. The workshop gives an overview and some practical hints for selecting a method without getting too deep into details of the methodologies. For visiting the course, it is not necessary to have knowledges in mathematics and statistics, further information on these topics is provided in additional voluntary videos on the platform. The workshop focuses on practical examples and typical challenges especially in the intercultural context. It is about how to do sampling by having in mind the challenges and difficulties, especially possible interferences with the interesting variables and correlations.

The course platform is composed of four parts: the first part is on non-probability sampling and discusses the different forms of purposive sampling used particularly in qualitative social research. In evaluations, there are many situations like the selection of cases, documents, interviews, members of focus groups etc where one must use such kind of sampling. The second part focuses on the mathematical background of probability sampling, leading to representative samples and offering the opportunity for drawing conclusions on population by using inference statistics. There is some information for those who are interested in the mathematical background, but merely the practical use will be discussed, especially when representativity is necessary. The third part offers examples from survey research, showing the practice by German private market and social research institutes. This is just one example of complex sampling designs with several layers and there are also well-established, comparable standards used by local institutes in the global south. As an evaluator, one should know about these standards and the difficulties implied in such procedures. Finally, the course will highlight on some examples from evaluation studies and the challenges for sampling in this context. Some key specifics will be mentioned as for example the intercultural differences that may influence response rates, the real world constraints in budget, time and data.

Key task during the course week is the development of sampling procedures and measures to avoid systematic biases. Depending on the number of participants, groups will be built, and they will work on the same example. During the online meeting, different solutions and their challenges will be discussed.


The participants …

  • know about the different selection problems,
  • have an overview on different sample approaches,
  • are able to apply key sampling techniques in evaluations,
  • understand the pros and cons of the various sampling procedures.

Recommended for

Evaluators and Practitioners






Read the detailed workshop description here.