We were asked recently to supply Age and Gender quotas for an urban-only population and we only had general population statistics to draw on. Data based on urban areas alone is not generally easy to find. So what would be the error if we used the general population breakdown instead?
We have a stereotypical view of people based on where they live. The centres of our metropolises are packed full of bright young things, hipsters with cool jobs and even cooler lifestyles, our rural areas full of people just “not like that.” So we imagine they are different and, when we travel to New York, to London or to Paris we see exactly the people we expect to see.
Is this true? And in the greater scheme of things does it matter?
To be mistaken and mislead by stereotypes is nothing new, it is part of the human condition. Generally we are not good with statistics (and this can even apply to market researchers).
Daniel Kahneman illustrates this for us in “Thinking Fast and Slow” with his question about ‘Steve’. An individual has been described by a neighbor as follows: “Steve is very shy and withdrawn, invariably helpful but with very little interest in people or in the world of reality. A meek and tidy soul, he has a need for order and structure, and a passion for detail.” Is Steve more likely to be a librarian or a farmer?
Most people immediately conclude that Steve is a librarian since he fits the stereotypical view of a librarian. They ignore that fact that there are many more farmers in the USA than there are librarians (by a factor of 5) and that many more librarians are women than men just makes the statistics even more in favour of male farmers over male librarians. Therefore of course it is much more likely that Steve is a farmer than a librarian, he just doesn’t look like a stereotypical farmer.
Why do we ignore the fact? It is both because we do not know this fact and because we don’t think it necessary in order to make an assessment. Researchers of course do not have this excuse. I looked up the number of farmers in the USA and the number of librarians and it was a relatively easy find!
So what’s the first question we ask ourselves about this rural/urban problem? Finding out how many people live in rural places would be a good start. What we do know is that the rural parts of a country are normally bigger, but most of this is taken up with fields, animals, mountains and the like. But what about the people? If there are very few people in the rural areas then, however different they are, they don’t make much contribution to the general population statistics.
In fact in the USA a little under 20% of the population is rural. A similar proportion is found in most industrialised, Western countries, the EU overall is 25% rural. It can rise to quite high numbers in emerging nations; Niger and South Sudan, for example, have 80% of their population living in rural areas. For the record I got that data from the World Bank.
So for the most part in developed countries the general population is 80% urban. And therefore the two populations (rural and urban) would have to be very different in order for the urban population alone to be dramatically different to the general.
As I wrote at the start such data are not easy to come by. The USA however does produce good small area statistics so we are able to make some comparisons. The table below is New York City compared to New York State in total and New York State excluding the city, which we label “rural”.
|Age||New York City||New York State||‘Rural’ New York State||USA|
New York City does skew younger than the state in total and the rest of the state. The average age of an adult citizen in the Big Apple is 46, in the Empire State it is 47. But then again be aware New York City makes up half of the population of the entire state. So perhaps the proper comparison is with New York State excluding New York City. Here the average age of an adult is 48. Yes it’s older, but only marginally so. Again also be aware that the majority of those outside New York City are also relatively Urban, we are talking about the cities of Buffalo and Rochester and even towns like Hempstead and Brookhaven. Eight cities and towns in NY have over 200,000 inhabitants – they alone make up over a third of what we are describing here as ‘rural’ New York State.
The final comparison is between New York City (standing here for all urban areas) and the nation. Yes there is an age difference, but it is marginal. The distribution by age groups is different, but only slightly. using either distribution for quotas would give you the same answer. Would I use the national numbers as proxy for the urban? I think I would.
So why should this not be obvious to all researchers? Perhaps because demography is one of our blind spots. We know it is there, just don’t always recognize how monolithic and slow-moving it is. And the hipsters on the street? Well that’s just a case of seeing why you want to see.