The May 2015 general election results have been described as a “debacle”, “terrible”, and a “disaster” for the very sizeable polling industry in the UK. At least ten major companies and one prominent private pollster produced regular polls in the run-up to the election. While not unprecedented (the general election of 1992 produced a similarly unpredicted result), the scale of error between what was estimated and the final result has been a major embarrassment for an industry staffed with highly qualified statisticians and researchers, and which often shaped the narrative and debate in the media during the run up to the election.
The tale of the UK election polling debacle is one that should also provide a wake up call for researchers, statisticians, economists and M&E professionals everywhere. Within international development in particular we have far less resource to collect data than pollsters, so when we do collect we need to be aware of where we can go wrong in our analysis. This LinkedIn post looks into what lessons researchers can be learned from the polling debacle….M&E, Research
Categorised in: Uncategorized