Crowds from high up_crop

OPINION24 March 2020

The demographics dilemma

Opinion Trends UK Youth

CrowdDNA’s Dave Power writes about the changing interpretation of demographic data among market researchers.

At Crowd Numbers, we are undertaking a diagnosis of quantitative research. As the most requested, and most frequently applied data approach, we felt that it was high time to think hard about where quant has been and where we plan to take it next. We are leaving no stone unturned, with the goal of reaching a point where we have sharpened our tools, cast out our bad habits, and prepared for the future. The first stone to turn over is a big one – demographics.

Demographics differences have long been the crutch of the lazy quant researcher. You needn’t shift your eyeballs past the first few columns of the data tables, and there they are. It’s like hunting for gold in a jewellery shop. “Men are significantly more likely to…”, “U25s and 35-44s are significantly less interested in...” It is simple eye spy stuff that we are guilty of dressing up as insight.

The recipient of these gems of wisdom is often left to convert this into something meaningful, and this can create dangerous biases. If the research identifies that a product is significantly more appealing among women than men, whose perception of a woman should we follow? And what is it about being a woman that drives that difference? It’s not clear, and this presents a problem. Depending on the cultural nuances of the market being researched (e.g. India vs. UK), or even the cultural backgrounds of the recipient, the researcher has lost control over the insight, creating an opportunity for the truth to derail.

And if we consider demographic reporting prone to misuse and idleness, then we must also acknowledge that the demographic frameworks we use for analysis are becoming increasingly frail.

Columns two and three in your table set are the most common crime scenes, where binary division of gender is still the standard way to cut data. But changes began in the previous decade, when it became commonplace to add additional options to gender selection questions, a change brought about by cultural pressure and shifting views.

Sixty per cent of Gen Z believe a binary option is not enough for a survey question, compared with 50% of millennials and 30% of boomers. But when it comes to analysis we fold people back into their biologically defined boxes. For now, this approach accommodates the majority of consumers, but what will be the best approach in 10-20 years time when Gen Z are the generation with the highest spending power? Should columns two and three still be simply cut as male and female?

Age breaks come next, columns four through seven; another stalwart of the beginner’s guide to quantitative analysis. For some time, we have understood that age is becoming less of a predictor for behaviour and with advances in health care and general wellbeing, our biological age is not so closely aligned with how we behave. On top of that, a number of cultural icons are shaking up our expectations of age.

American model JoAni Johnson who at age 67 became the new face of makeup brand Fenty (having made her runway debut aged 65 ); Baddiwinkle, a prolific 90 year old influencer, preaches to 3.8m followers on Instagram; and Inge Ginsburg, a holocaust survivor, fronts a death metal band in Switzerland at the age of 96.

But then at the same time, major issues have illustrated the growing disparities between generations, as highlighted by issues such as Brexit, the Hong Kong riots, and the aforementioned views on gender and identity. So if age both can, and can’t predict our behaviour, how stable is this as an analytical device?

One generation in particular is presenting itself as somewhat unpredictable –  Generation Z. Our research frequently identifies an endless number of hybrid states that coexist with one another, pulling their views in opposite directions on common topics. For example, Gen Z may have held strong and progressive opinions on gender identity, but at the same time maintain very traditional views on a subject such as marriage.

They can be avid supporters of universalism, but firm believers in individualism. They can express moments of true anxiety (environmental factors) and moments of genuine optimism (career). So when analysing data collected from Gen Z we must take extra care to consider the wider cultural context and whether a finding typifies Gen Z, or highlights another hybrid state.

For now, demographics still have their place in our analytical tool kit, but they are coming under pressure to change, and this pressure will likely grow at an exponential rate. Brands still wish to understand them, though many no longer want to market to them. Media brands still use demographics as a common currency for their audiences, but then editorialise the dissolution of their boundaries. Insight professionals still record and cut their data using them, but it’s time we consider what we will do next.