On Sunday I wrote a piece on Ofcom’s Diversity in Television report, and in particular, noted my disappointment that it didn’t measure social class.
The feedback I got can basically be summed up with the question: “Yes, but how do you measure class?”
So I thought it was worth exploring the issue a bit further.
Measuring social class isn’t easy. What you can’t do is simply ask people to mark themselves on a form. You need to collect proxy information that can provide you with some kind of methodology to measure it.
Here we come to census v survey.
A census is a record of every single employee, whereas a survey is a sample of some of the population. While ordinarily you’d want to measure the responses of all your employees, if your company is big enough then a survey may suffice. Not only that, if you know that some employees are likely to feel uncomfortable answering certain questions, then you’re likely to need to use a survey.
It’s for this reason, by the way, that surveys conducted about sensitive areas such as sex, should be treated with extreme caution, since many do not wish to answer, and indeed may be answering untruthfully.
Of course, there are rightly concerns that this is sensitive data. What right does my employer have to know about my parents’ education, or jobs? And as an employer, do I feel comfortable asking employees to collect this data?
It is sensitive information, and it needs to be collected and measured responsibly. So that probably means that it shouldn’t sit as a field in an employee’s record on an HR system, anymore than you’d record someone’s sexual orientation or religious beliefs on such a system.
Yet we also collect data on those sensitive areas. It’s usually collected in survey form, and on an anonymised basis. The collection is probably best handled by a third-party specialist research company who can assure employees that the data is not being used for anything other than measuring diversity in the workplace.
It’s important that social class data is collected as it impacts on many behaviours across societies. So while it’s hard to do it, groups like the Office of National Statistics have to collect this data, and indeed they have their own methodology for doing so. Notably, these are based around employment status (employer, self-employed or employee), organisational size and supervisory status (does a person supervise others, and in what context?).
As The Guardian reported over the weekend, the BBC has made the decision to use a staff survey which measured parents’ occupations, noting that its staff showed a higher likelihood of their parents having achieved higher managerial and professional occupations than the wider population, suggesting a class imbalance compared with the wider population.
Now it’s certainly true that an organisation the size of the BBC is able to get an external research company to measure such indicators, and provide norms to compare against. But Ofcom’s report was based on UK broadcasters who all had turnover’s of £1bn or more, so I’d argue that each of them is in a position to do a similar job.
On the other hand, a small indie isn’t in such a position, and the size of that indie might make such data relatively meaningless anyway.
Yet if the media industry is serious about diversity, then this does need measuring, and doing so on a pan-media basis with some central funding, could mean that the broader industry could be surveyed.
Mind you, as a friend of mine said to me, if you banned unpaid “internships” tomorrow, it may fix the problem quite quickly.