LITERALLY nothing is more popular than Taggart on Scottish television.

Or so it would appear from new audience data that shows a blank screen is one of the most watched shows on STV's local channels in Glasgow and Edinburgh - ahead of three episodes of the popular Scottish detective series.

STV has boasted that its City TV services, STV Glasgow and STV Edinburgh are delivering an audience with an average reach of 759,000 viewers per month, 30 per cent of the audience in the transmission area.

Now it has emerged that the second most watched broadcast in one week on STV's hyperlocal channels was when 9000 saw "This is STV Glasgow" in the early hours of one Sunday morning, which features a screen saying there was no programmes on yet.

The audience figures for the week ending March 19, shows that's 1000 more than watched an evening episode of Taggart later in the day.

The Herald:

Eight thousand had tuned into STV's equivalent of the test card in the early hours of Friday morning, and a further 7000 were watching in the early hours of the previous Wednesday.

Only an edition of Peter and Roughie’s Football Show, a televised phone-in programme aired on the Wednesday did better than Sunday’s 'nothing to see' message - which normally runs when there are no scheduled programmes, between 2am and 8am.

It has led one TV analyst to believe that the bizarre ratings may have uncovered a flaw in the gathering of data by the Broadcasters Audience Research Board when dealing with small audiences.

The Herald:

BARB make their calculations about TV audience across all networks using a panel consisting of 5100 British households and has becomes the only yardstick for assessing how programmes, channels or marketing campaigns have performed and provides the basis for airtime advertising trading.

The new figures led to speculation that the spike in audience for Sunday morning's This is STV Glasgow may have been brought about by the small number of Scottish panel members who may have had the hyperlocal channels on, falling asleep in front of the TV after a Saturday night out.

The Herald:

Once a household has been recruited to the BARB panel, their home is fitted with a meter on every TV, laptop, desktop computers, and tablets.

In order for the meter to know who is watching, each member of the household over the age of four is assigned a button on a special remote control.

If they enter a room while the television is on they must press their designated button to register their presence and press it again when they leave to show they are no longer watching.

One Scottish television insider said: "It would appear to be a baffling anomaly. It might be that the eventual sample size for local TV is too small to accurately gauge audiences at that level, although there will be some who will leap on that and say it may reflect lack of interest.

"Generally speaking the margin of error in BARB audience figure is acceptable to advertisers and broadcasters when it comes to major networks, but what the question is whether it works when it comes to channels with a small audience."

BARB will not discuss how many Scots are recruited for the panel, let alone those who might be watching STV City, and could not offering any definitive explanation for the spike in the 'no telly' viewing.

The Herald:

But one BARB source said: "That is interesting. It is possible that a number of panel members were watching TV and left it on while the test card was up. I don't think one panel member could have driven in up that far.

"We are continuously looking at where we need more panel homes to make sure it is proportionate throughout the UK and then it is weighted and measured accordingly."

It's not the first time BARB data has left broadcasters and programme makers scratching their collective heads.

Three years ago Ron Jones, the founder and executive chairman of Tinopolis, one of the largest independent TV production froups in the UK, commissioned a statistician to analyse the way TV audience figures were measured in Wales, saying he believed the method may be inaccurate.

He said he found it "very odd" that BARB igures for S4C programmes produced by Tinopolis suggested that as many as 10-15% of viewers were taking part in competitions promoted as the shows went out, when the normal proportion was under 1 per cent.

Addressing the peculiar Scottish data, a spokeswoman for STV said: “Clearly there is an anomaly in the figures. STV’s City TV services, like local TV service across the UK, are currently working with BARB to develop an appropriate measurement.”