★★★★★ 0/5

Ratings investigation over overlap audience

Have a group of Sydney metro viewers been miscategorised as regional viewers?

10
Leave a comment

An investigation is underway following questions whether a group of Sydney metro viewers has been miscategorised as regional viewers.

Data is collected by Nielsen on behalf of OzTAM, owned by Seven, Nine and 10.

According to The Australian, questions date back to June, and involves what has been dubbed an “overlap” between Sydney’s southern suburbs and Wollongong, which is considered a regional area for ratings purposes. In many of these southern suburbs, it is possible to view both Sydney and Wollongong TV stations.

“A small percentage increase has been identified in viewing out of the metro Sydney market to network affiliates in the ‘overlap’ area in Wollongong since early June,” a spokeswoman told the newspaper. “Nielsen is investigating whether this is accurate or needs attention.”

Sydney and Melbourne both have 1475 homes in the OzTAM a=panel, 1000 in Brisbane, Adelaide & Perth 650. Regional areas are Queensland (812), Northern NSW (700), Victoria (651), Southern NSW (570), Tasmania (285) and Western Australia (180). A Total TV database (‘VOZ’) includes a ‘Rest of Australia’ market that captures areas outside OzTAM and Regional TAM markets.

10 Responses

  1. I do see this as an issue that is coming forth as the major metropolitan areas sprawl or towns and villages within commuting distance of the metropolitan areas acquire more people. This would be underscored whenever a motorway-grade road and/or a faster railway is built to connect a satellite city to the major metropolitan area thus shortening the commute time between these areas or metropolitan-area employers are encouraged to set up shop close to the periphery of the metropolitan area.

    It would encompass more people living in what were two different TV / radio market areas and able to receive stations within both areas. What I have seen happen with London is the powers that be who define the market areas or licence areas redefine areas like the main London TV/radio market area to encompass the cities and towns making up London’s commuter belt. Or it may be about reclassifying large regional cities like Wollongong, Geelong, Gold Coast or Newcastle as their own “metropolitan…

  2. This issue dates back to the 60s. There’s a stronger Wollongong signal into Sydney’s southern suburbs than a Sydney signal, due to terrain. At the same time we get a stronger Sydney signal into Wollongong/Illawarra than the ‘local’ Illawarra signal. The weakness of the Illawarra signal is due to the original location of the Knights Hill tx south of W’gong whereas Sydney tx are north of W’gong and unobstructed. The Illawarra escarpment blocks the local signal in many coastal areas of W’gong. There are a couple of translators to serve Wollongong suburbs, very low power (10w-16w) and directional. The Knights Hill antenna is directional to the south. Homes on southern side of Lake Illawarra can’t receive any signal from nearby Knights Hill but can receive Sydney perfectly. Again, the escarpment and Mt Warrigal blocking. When TCN9 blacked out half of the day’s cricket Sydney viewers turned to WIN for the full day’s play. Why many antennae in ‘The Shire’ still point south to…

  3. The last paragraph containing details of the number of households involved in the ratings is fascinating. It’s extraordinary that they can get a statistically viable sample out of less than 0.1% of the population.

    To be honest I don’t know if it’s extraordinary in a good way or a bad way. Certainly raises questions though.

    1. Yeah it’s low but seems to work as in Metro is 5,600 OzTam homes to 9.85 million TV Homes at 0.056%, whereas in the USA it’s 20,000 Nielsen Households to 128 million TV Households at 0.016 (3.5 times less). At least it’s just TV Ratings and while important not country altering, as generally all those Political Polls we see here (and in the USA) are based on surveys of a lot less and they can end up deciding far more.

      1. *And yes I got that incorrect, as the 9.85 million homes is Aust wide not just Metro so it’s more like 0.0893% of Aust homes not 0.056% (8,798 vs 5,600), so we have an even better sample than in the USA (just over 5 times better).

    2. It’s just statistical sampling theory. If you sample 2000 whose demographics match the population closely, the data will be accurate to about 2%. It doesn’t matter how large the total population is. Which is why the US can get with a small sample and large population. It’s exactly the same way election polls work. The problem arises when your sample drifts away from where it is supposed to be. As happened when Oztam took over from Neilsen and Nine’s ratings were way out.

      1. Thanks for that, I knew they were accurate apart from the odd anomalies here and there like you say with Nine and of course the Australian Election when it was Shorten v Morrison and the US one of Clinton v Trump, but wasn’t up on statistical sampling theory (now searched and read about). In the US Nielsen got their accreditation *suspended back in 2021 and it still is due to under reporting as well, however that was more to do with covid time watching errors and they have been updating figures. Of course that didn’t stop Nielsen being **sold to the Evergreen- and Brookfield-Led Consortium for $16 Billion back in March, again anomalies happen but in the main it’s been accurate hence the sale price.

        *thedrum.com/news/2021/09/01/nielsen-mrc-accreditation-suspended-following-accusations-underreporting

        **ir.nielsen.com/news-events/press-releases/news-details/2022/Nielsen-Enters-into-Agreement-to-be-Acquired-by-Evergreen–and-Brookfield-Led-Consortium-for-16-Billion/default.aspx

Leave a Reply