Abstract:

In this paper, I want to explore the best ways of measuring basic demographics (age, gender, education, income, working status, job and socio-economic classifiers), based on two years of exploratory research-on-research, comparing different methods used around the world and testing new techniques. These demographic questions, asked in-survey, are the bedrock of market research. We rely on them to target audiences, set quotas, purchase sample and ensure our research is representative. There are, however, no international standards for asking them and few established best practice. Everyone has been left to their own devices, with methods varying dramatically from one market research company to the next. In addition, many countries have their own way of measuring social class and job status, to the point that it can be extremely difficult to compare cross-market data. Moreover, anyone wishing to mix and match sample from different supply sources can rarely be confident they are getting comparable spread. Perhaps worse still are the issues of data quality and bias. Over the last few years, we have explored the gap between a range of survey answer data and what we know from elsewhere to be the truth, in an attempt to understand the underlying biases that affect data. Sadly, some of the largest gaps can be observed in basic demographic questions. This has been an elephant sitting in the room of market research for several decades; it might be one of the biggest inefficiencies holding back our industry.