Blair Jenkins of Yes Scotland has said that the BBC was not biased in its reporting of the referendum campaign.[1] For my part, I believe that much of the criticism aimed the way of the BBC is misguided, in the sense of being misdirected; that’s not quite the same as saying that the criticism is all wrong, however.

Some years ago, when my son was still a teenager, he and I watched a TV programme on the Miners’ Strike of the 1980s. He was fascinated, particularly by the scenes of the massed rammy with the police which has since come to be known as the Battle of Orgreave. That this kind of thing could have happened in modern Britain was something he hadn’t been expecting.

What made almost as much impression as the visuals was the commentary added to the programme. Whilst most of that was a recent voice-over, looking at things from today’s perspective, some of it dated from the period itself. In that original commentary there was no attempt at striking any balance; instead, the miners were hoodlums and hooligans whereas the Government was simply reacting to an issue of public disorder.

My son’s reaction took me back to when, as a student myself in the mid-1970s, I came across the work of the Glasgow Media Group.[2] This was a revelation for me at the time: it’s not that I hadn’t wondered previously at the content of news reports, but what I hadn’t seen before then was anything which gave expression to that questioning and sought to explain how these things work.

Being young and foolish, I made the mistake of pontificating on the subject to my own Dad. He was having none of it: “the news is the news; how can it be biased?”. Armed with my new-found expertise I asked him to tell me when it had ever been that a Royal Visit had been reported as causing disruption to traffic and city centre businesses, in the way that trade union demonstrations always seemed to be doing, and how it was that TUC leaders were always “barons” whereas CBI bosses were “captains” of industry. (In case you’re wondering, I didn’t succeed with my argument. It was a tactical error I regretted for a while afterwards.)

To a statistician, “bias” is a possible feature of any sampling or estimating exercise. It will occur, for example, when your sample does not match the overall population it was meant to represent. That may be unavoidable given the limitations of the data collection logistics (e.g. the cost involved or the time available). If your shopping centre interviews can be done only on Tuesday afternoons then you’re not going to come across too many young teenagers (and those you do encounter will be a of particular type, rather being the norm for their age group).

Pollsters are well aware of this and adjust their results in order to correct for such bias: if there aren’t enough older women in the sample to match the overall population, you give more weight to the ones you do have to boost the contribution from that sector.[3]

There is a whole range of types of bias, reflecting the various ways it can be introduced. One worth highlighting here is that of the estimator. Let’s say that, when analysing your polling results, they are different to what you were expecting and just look wrong. There is then a temptation to find reasons (or excuses) for that and to use those as justification for adjusting the results to bring them back into line. The equivalent is present in a great many fields of activity – just look at the reactions to a party leader’s speech, which differ quite predictably according to the prejudices of the audiences concerned (which is a form of observer bias).

Some bias we find acceptable. Think of the example of a football match, which is later shown on TV in an edited form, ninety minutes reduced to only ten or perhaps twenty. Comparing the two, what you’re likely to find are the following:

• lengthy periods of boring, uneventful play are omitted (highlights, after all)
• flash points (disputed decisions, bad fouls, offsides, penalty awards) are dwelt on for perhaps 20% of the coverage whereas, as the time added on suggests, they in fact took up at most 5% of the actual match
• if it had been a particularly one-sided game, it will appear far less so in the edited version (in the interests of balance . . .)
• and there’s many a match where, for those in attendance at the stadium itself, the crowd has been almost as much of an entertainment as the players, which gets lost in the translation to TV (unless there are Brazilian beauties involved, of course).

In the main, we don’t take issue with that; it’s expected to be that way and no-one feels short-changed. There are occasions, however, when controversies do arise: the selection of dodgy decisions showed all the claims against our lads, but ignored lots of blatant cheating by thae other b*****ds. More serious, the broadcasters routinely tune out singing by Old Firm crowds that most would find unacceptable, on the grounds I assume that there’s a consensus that we prefer not to be subjected to that bile in our living rooms.

In terms of trying to maintain objectivity, the most challenging example is the “in-depth report”. The production team spends a period gathering film footage on the subject. This is then edited down to fit the length of the broadcast slot available and a script produced for the accompanying voice-over (or perhaps the other way round). The act of creating this narrative cannot help but be influenced by the political and moral viewpoints of those concerned; the choice of footage and the language used will betray that. Anyone who tells a story has some point or other to make, whether conscious of that or otherwise.

The point is that all editing decisions are examples of bias in operation. It is important to keep in mind that such bias is present, even if we find ourselves content with the results. In all circumstances, any adjustment at all to the source material – of poll data or of raw film footage – involves making a judgement call, which by its very nature has to be open to a bias on behalf of the person concerned.[4] Even the choice of source material to gather in the first place is itself open to selection bias.

What I remember of the TV coverage of the Troubles in Northern Ireland is that it seemed even-handed enough to begin with (more perplexed by the situation than anything else) and I would say that it has ended up fairly so in recent years (now that the temperature has subsided following the Good Friday Agreement).

The years in between were different, though: the language used didn’t always seem to be impartial in tone.

That should not be a surprise, of course, given that during this period the atrocities were extended to include the mainland itself. Devastating bombs went off in British cities. There was even a mortar attempt on Downing Street.

When the State feels itself to be under attack, do you not expect it to fight back with all that it has at its disposal, including mass communication? Weren’t the newsreels of WW2 used in this way as well?

On 19th August 2014, Permanent Secretary at the DWP Sir Robert Devereux issued a memo to civil servants telling them that, whereas the Civil Service had to remain neutral when it came to a UK General Election, there was to be a different approach for the referendum: “the UK Government has a clear position to maintain the Union and so it is legitimate and necessary for UK civil servants to support the Government in this objective”.

You can understand the argument he was making, but the issue with it is that there is in fact no such established constitutional principle that backs it up, which means that the point is moot at best and invented at worst.[5] The UK Civil Service – supposedly a vital component of the system of “checks and balances” – was instructed to put its shoulder to the propaganda wheel.

And, no, “propaganda” is not hyperbole. It’s part and parcel of what all governments have in their armoury when circumstances require it. Paul Mason’s view appeared on Facebook during the later stages of the referendum campaign: “Not since Iraq have I seen BBC News working at propaganda strength like this”.

Aside from commenting on the then current situation, what Mason was confirming, albeit indirectly, was that at times of crisis the BBC will be brought into the propaganda effort. When I think back to the Falklands War, the Gulf War, Afghanistan and Iraq then the truth of that becomes clear. Added to that list should of course be the Miners’ Strike.

It may have passed most folk in Scotland by, but the biggest complaints about BBC bias in the recent past have come from UKIP. A bit of googling will throw up plenty of examples of that. Nigel Farage used to be presented as a figure of fun, treated only with ridicule and disdain. BBC reporters barracked him. All that has changed, of course, with there being coverage a-plenty, now that Nigel’s Barmy Army is feared, having become a serious threat to the established political order.

Remember the appearance of Nick Griffin, then of the BNP, on the BBC’s Question Time? There wasn’t even a token pretence at impartiality by the chairman, David Dimbleby, who cut Griffin off repeatedly, did little to stop others talking over him and allowed heckling of him by the audience. It was as slanted a session as I’ve seen. You might not regret that at all, but let’s not kid ourselves about the facts here: it was an example of blatant bias.

The BBC is an organ of the Establishment. When it is called upon to do so, it acts to defend the Establishment. It has become clear also that it will do so without needing to be called upon, as it is its default stance. Via Sir Robert Devereux, the Civil Service has admitted that it is not impartial when it comes to threats to the established order and the same is true – whether admitted or not – for the BBC.[6]

The BBC is driven by the desire to seek a balance, but its blinkers mean that it has always viewed that in the simplistic way of giving equal representation to the political parties. That’s acceptable for many circumstances, but it falls down when it comes to single issue matters; instead of balance in terms of the issue in question, it gives an equal voice to each of the main parties. When most of the parties line up on the same side – as happened with the referendum – it results in an imbalance as regards the coverage of the issue. The same will apply when it comes to the promised EU referendum; just watch Farage moan his face off about that.

Is the BBC biased? Of course it is, but that’s not the real issue. There is bias everywhere.

Alex Salmond said this in a Channel 4 interview: “There’s a huge difference between being a public service broadcaster and being a state broadcaster, and I’m not certain that the BBC understand that difference“.

That’s the issue. The BBC can’t be both at the same time, even if pretends otherwise. There’s no point in bleating about that, though. You simply have to learn to deal with it and to play ‘em at their own game.

The likes of the BBC are not the target; its audience is. You should be looking to cut through the institutional bias, seeking to exploit it so that the arguments are seen by that audience to be falling your way, enhancing your position. A bias cut, so to speak.

 

 

Footnotes

1. Interview with STV News, as reported on STV’s website:

“I think there were mistakes made by the broadcasters, and I think there were omissions and I think sometimes because broadcasters are not always as well resourced as once they were there is a tendency to pick up the newspaper agenda unthinkingly more than was perhaps the case in previous decades.

We had issues from time to time, but I don’t myself (think) that we faced a systemic bias if you like, that there was some corporate intent to disadvantage the Yes campaign.

Knowing how both STV and the BBC operate, and knowing a lot of the people involved I don’t think that’s a credible view.”

2. Glasgow Media Group, 1976:

“Contrary to the claims, conventions and culture of television journalism, the news is not a neutral product. For television news is a cultural artefact; it is a sequence of socially manufactured messages, which carry many of the culturally dominant assumptions of our society. From the accents of the newscasters to the vocabulary of camera angles; from who gets on and what questions they are asked, via selection of news stories to presentation of bulletins, the news is a highly mediated product.“

3. From a statistical analysis point of view, doing that introduces a risk of bias of a different kind. If the older women you do have were for some reason not representative of their population sector as a whole (you just happened to have run into a church group making its way through the shopping centre coming back from a coffee morning) then you will have skewed the overall result as a consequence. It’s why it is that the job of a pollster is more complex than might be obvious at first.

4. See my earlier post regarding the Nick Robinson controversy.

5. There is of course no written constitution at all. There are rules governing the conduct of referendums in general in the UK and in the particular case of the 2014 referendum there is the Edinburgh Agreement; I’m pretty sure that this particular provision does not feature in either of those.

6. We will need to wait to see if some disgruntled insider does eventually leak a BBC memo on editorial policy during the referendum.