Sampling

When conducting research from a population of interest, a well designed sample enables you to find the right balance between quick, accurate and cost effective answers. If a sample is too small, then estimates will be imprecise and the conclusions which can be drawn from it will be limited. If it is too big, then the sample will be time consuming, costly to implement, and place unnecessary burden on respondents. And if it is not suitably representative, then it will be difficult to apply its findings to the real world.

By working closely with research and survey administration experts at NFER and with our external clients, NFER’s statisticians are best placed to find this balance. By understanding the research need, we are able recommend the most appropriate sampling design and size, ranging from simple random samples, to more advanced techniques such as stratification, weighting, multistage and clustered samples. We can also develop and implement protocols for sampling and randomisation as part of Randomised Control Trials (RCTs).

Please contact Jo Morrison if you would like to discuss how NFER could help you with sampling.


Statistical analysis home