Simple random sampling has traditional been the mainstay of market and social research surveys. Obtaining a simple random sample has become even more elusive in an on-line world where community panels and convenience sampling prevail.
Moreover, there continues to be increasing difficulty in accessing sampling frames for the population of interest as well as a continuing decline in response rates. Nonetheless, survey research has proven to be resilient and adaptable in the face of these challenges.
This Seminar will adopt a solutions-based approach, illustrated by case studies, which show how inferences can be improved from surveys administered to biased, low response rate and non-probability samples.
In so doing, it will address how to improve the accuracy of the survey estimates we generate from poorer quality and non-probability samples.
The new reality - low response rate probability surveys
Advances in survey execution and weighting techniques to adjust for non-coverage error and reduce non-response bias in low response rate probability-based surveys.
Advances in techniques to improve estimates from non-probability samples illustrated by case studies in the areas of sample blending, calibration and weighting