When Statistics Hide The Truth

I recently had dinner with a friend who works on public health issues in the developing world. During the course of the meal the conversation ranged over a number of different subjects and I innocently asked her whether health rates were getting better or worse in Africa.

She’s known me almost all my life and gently told me that my question was a stupid one with the following response;

“Getting accurate statistics is almost impossible but even if you could often they don’t tell you the whole picture. Take a city like London, are Londoners healthier than they were twenty years ago? In some parts of the city the expected mortality rate is in the 80’s while in other neighbourhoods it’s lower than that of Malawi. Knowing the average mortality rate for a big city normally hides more than it reveals. If I’m going to improve a country’s health I don’t look at one big picture I look at loads of small pictures”

That warning against using big overall averages is one that I think anyone interested in diversity in television needs to listen to.

According to the big average official statistics when it comes to the number of black and ethnic minority people working in television things seem to be great, they might not be perfect, but the BBC and the broadcasting industry generally appear to be moving in the right direction and reflecting the diversity of broader society.

For instance, the percentage of the UK population that is BME is approximately 13%, while the percentage of the BBC that is BME is 10%. The percentage of the general broadcasting industry that is BME is 10.4%. Overall, it’s not perfect, but it’s also not too bad, particularly compared to other professions and industries.

The problem with the official statistics, however, is that they seem to fly in the face of my everyday reality. I constantly go to meetings where I am the only non-white person at the meeting. Recently I was at the Edinburgh Television Festival and, the festival’s participants were overwhelming white. The idea that 10% of the broadcasting industry are BME seemed fanciful in that context.

So how can these two facts possibly exist simultaneously?

A few years ago, a similar dilemma faced Professor Becky Pettit from the University of Washington. When she looked at official surveys conducted by the American government, African Americans seemed to be doing far better than the on the ground reality she was seeing. As she sets out in her new book, 'Invisible Men', her breakthrough came when she realised that many of the surveys excluded people in prisons. When the prison population was included the picture of how well African Americans are doing was dramatically altered, to be more in line with the reality she was experiencing.  That makes sense – African Americans are a whopping seven times more likely than the general population in the USA to be incarcerated.  The official statistics just didn’t reflect the reality of African Americans. (To use my dinning friend’s analogy it’s as if the neighbourhood with the worst health statistics had just been left off the official numbers)

So is there a simple accounting error that could be giving the wrong picture of how well BME people are doing in the television industry?  Would that explain why my day to day reality seems to be so different from the official statistics?

I suspect that there isn’t one big factor such as the one uncovered by Professor Pettit, but a number of different factors might, taken cumulatively, create a skewed picture.

Here’s one factor.  Different broadcasters define BME in different ways. For example, some broadcasters define BME as anyone who is not “white with a British passport”. This means white Europeans (including white Irish), white Americans and even white Australians and New Zealanders can be included in the official statistics.

A second factor might be that often freelancers and people working for independent television companies are excluded from the statistics. As the television industry increasingly becomes more reliant on freelancers and indies it is very hard to describe any of your statistics as definitive if you are excluding such a large number.

And last but not least, a third factor highlights my friend’s concern of using large average numbers. While the official statistics have now begun to make a distinction between different grades – a good move – there is often no distinction made between where someone works for a broadcaster. This means that production roles with editorial responsibility are effectively seen as exactly the same as jobs with no editorial responsibility.  And this could be one of the reasons why I am the only black person at production meetings. Many other BME people are employed in important positions but, often, they are not involved directly in editorial decision-making. Knowing the average number of BME people working at the BBC or ITV or C4 does little in telling me how well BME employment is in News and Current Affairs, Drama or any other genre.

Despite the big average statistics, my reality suggests that we could all do better when it comes to diversity.  What is your reality? Are the big average statistics hiding the more important smaller truths?