Rachel's blog

Co-founder of Throng. Favourite TV shows recently: Homeland and The Newsroom.

Stuff’s article What’s Wrong With Television may be an interesting read with plenty of things to discuss, but there’s a number of statements in it which the data on television viewership in New Zealand simply doesn’t support:

“There are many reasons why people are switching off their traditional television…people are also turning off in droves because…It is no wonder, then, that people are turning their televisions off…Many of us are also forgoing watching shows on television at all…”

Contrast that to the latest data from Nielsen’s:

“In 2011 New Zealanders spent more time than ever watching Television.  While the average person’s viewing time of 3 hours and 22 minutes per day remained unchanged from 2010, more people tuned in each day, resulting in growth of total hours tuned of 2%. Overall 2011 saw an additional 55,000 people tuning in, with 75% of the population now watching Television on a daily basis. Over 3 million New Zealanders now watch Television each day.”

By all means discuss the future of broadcast television, but the article gives the wrong impression about viewership trends and figures.

I’ve been asked to be on the panel for this week’s Media7 on a discussion about the public broadcasting service.  

One of the things they were interested in were my thoughts on the statistics quoted by the NZ Herald in two editorials:

“[TVNZ7] attracted just over 200,000 viewers a week. That number could never justify the station’s $15 million cost… This really is television for minority interests taken to extreme. By way of contrast, TVNZ’s One News attracts 600,000 viewers each and every night.”

Russell Brown writes that that the “just over 200,000 viewers a week” figure in the column “does not have this right”.  Unfortunately he does not provide the corrected figure for the weekly cumulative viewership but instead quotes a different measure that is usually reported for channels like Maori Television and Stratos: the monthly cumulative viewership.  

“The average monthly cume for TVNZ 7 in 2011 was 1.1 million. The channel finished the year with a record 1.4 million viewers in December, according to Nielsen.”

I like that Russell has quoted the statistic clearly: he’s provided the time period (average for 2011 and December 2011) and the source (Nielsen).  

In contrast the Herald does not provide the time period (for example, were the 200,000 viewers a week from the low summer period?) or the source so it can be verified. 

I’ve been asking Media7 for the average weekly cumulative viewership for TVNZ7 in 2011 so we know what it actually is.  Even if the weekly cume is not what they normally report, it would be good to clear this up rather than try and compare apples with oranges.  

Comparing a minor channel’s weekly viewership with the highest rating show on the biggest TV channel is not a fair comparison and clouds your judgment about what exactly is a “big” number.

Let’s just assume the Herald is correct.  200,000 is a big number.  It’s about the size of the crowd who turned up to Auckland’s Christmas in the Park recently.  Is that big crowd what you’d define as a “minority interest”?  And this is the number turning up each week to TVNZ7.  200,000 is about is about 1 in 20 New Zealanders tuning into TVNZ7 in a week. 1 in 20. 

I’m curious as to how many people does the channel need per week to justify its funding for the NZ Herald to be satisfied? 

The taxpayer funds plenty of other arts programs and far less people are turning up for those.  For example, 4.4 million was given to the NZ Ballet in 2010 and 69,000 people saw them at some stage during the entire year (and paid for the privilege).  

So it can’t just be about the numbers.  When reading the thoughtful book Monoculture, I was reminded that government has evolved to run more and more as a business and things created for the common good have been devalued.  

I don’t think there’s been enough thoughtful discussion about the future of public broadcasting in New Zealand.  It doesn’t appear to be in the media’s interest to discuss it in a non-partisan way since they are in competition, nor in TVNZ’s interest to have had TVNZ6/7 succeed: more viewers for those channels would be at the expense of their commercial channels.  There hasn’t been much of a political fuss either.  The most vocal campaigns have been via Facebook groups to save Kidzone and TVNZ7.  

I’ll be sad to see the end of TVNZ7.  I still miss Kidzone.  

I may not tune in each week but I support there being a place for Kiwis to learn their craft, share their stories, our accents and our culture and it not being behind a paywall.  


Did you know that people interviewed on ONE News and 3 News are referred to as “talent” by reporters?  We were surprised and uncomfortable to hear it being used in this context.  Of course, we’ve heard it in relation to actors, hosts, reality show contestants etc before.

When Kate Lynch and Amanda Gillies clarified their recent stories to Throng, both used the curious word:

We weren’t the only ones to notice the use of the word.  

Vaughan Davis tweeted: “‘Talent?’ I’d never heard that term used for the subject of a news story before. Is it common?  Familiar from ad and programme production but hadn’t heard in a news context.”  

Wendypoo tweeted: “I was a little gobsmacked at that term too!  Yeah I wouldn’t have thought it would be used in news terms, doesn’t sound like a very creditible term”

What do you think about the use of the word “talent” in the context of a person being interviewed by a TV news crew?  

I took this photo of TV ONE around 11:30pm tonight:

I’ve been following along also on the Electoral Commission website and it says that the result is a tie between Nicky Wagner and Brendon Burns.  All the other news sites I’ve found are saying it’s a tie as well.  Has TVNZ got this wrong?

“I don’t think anyone here has ever watched TV3 to be honest, it’s a new thing for them tonight.”

– Rachel Smalley at NZ First Party Headquarters

I just saw this graph on TVNZ’s election coverage: 

Putting aside the issue of using a perspective graph makes it harder to compare, there’s something wrong with this graph: the informal votes (labelled INF) bar is not correct.

The correct graph should look like this:

Update: there are faint notches in their graphic indicating the “0” mark, however it’s not a clear enough distinction.

Tonight I was invited along to the final leaders’ debate between Mr Key and Mr Goff at TVNZ.   I arrived at 6:15pm, and mingled with people over nibbles and drinks.  There was a decent mix of people there and lots of young adults too.  A group had a discussion about whether TVNZ’s Christmas tree being decked out in just blue and green decorations was meant to be a “subversive/meaningful” thing or not.  Funny!  

The Prime Minister walked straight in past us all but didn’t stop to talk.  Mr Goff walked past and then came back briefly to chat, talking about maybe having a beer with people afterwards.  He seemed very relaxed.  

We then went through the rabbit warren in TVNZ – past the news room (I always love walking past there), past the rooms where Mr Key and Mr Goff were waiting (with their minders with earpieces hanging around outside the room) to sit in the studio audience behind the politicians.  

There were about 9 crew in the room that I could see and maybe 6 different cameras or so.  It’s been a while since I’ve been around a live TV show being filmed. (The last TV show I saw being filmed was Hawaii Five-0 in August.)  I’m always struck at the magic of TV making it all look so good for everyone at home and all those hard working people behind the scenes making sure everything runs smoothly on screen – floor managers, makeup, cameramen, directors, sound guys, etc.  

I was one of the last to enter, so ended up being in the front row very near the center so you can see me behind the Prime Minister in most shots.

The floor manager told us we were to be a neutral audience – no clapping/cheering/sniggers etc.  (The audience did clap them both at the very end, off air)  The only noise we were allowed to make is a laugh if one made a joke 🙂

We couldn’t hear what Mark Sainsbury was saying (he was upstairs at TVNZ) but could see him on a little monitor.  Guyon Espiner’s autocue was just beside me – I’d turn during the ad break to see what he was going to say next.  Of course, most of what Guyon said wasn’t on autocue – but the intros at the beginning of each segment were.  

The people around me were all very friendly and excited to be a part of the audience.  Some had been in the audience for last week’s debate with the minor party leaders, but others had never been into TVNZ before.  

One of the things which struck me was how strange it was to hear Mr Key and Mr Goff’s voices not through a microphone.  I’m so used to hearing them being very loud or through a microphone that their voices sounded different somehow to me.  Even though it was a big event, it felt so much more natural hearing their voices like that.  

During the ad breaks, Mr Goff and Mr Key would come and talk to people in the audience.  I talked to both.  I asked Mr Goff how much sleep he was getting in the last week of the campaign.  He said 5-6 hours normally but that he power naps on the plane and falls asleep even before the plane is on the runway.  He told me he is known to curse under his breath when reading what the papers have to say about himself but he’s learnt not to let criticism bother him.  He was very jovial and joking, laughing lots with people in the audience in the breaks.  

I found it funny how Mr Key had his feet under the chair – most of the time he had one foot curled or his toes hooked around the base of the desk chair.  It reminded me of kids at school doing that.  

The Prime Minister asked me what I did, so we talked for a whole ad segment about my web design business and broadband.  Funnily enough, he used me as an example in a later segment in the debate!  Straight after that segment, he made a beeline for me and said he hoped I didn’t mind being mentioned (I didn’t!) and we talked a bit more about my boys and enjoying the chance for a night out tonight.

That was the fourth time I’ve met a Prime Minister in office (I had also met David Lange, Helen Clark and Jim Bolger) and it was the time I’ve felt the least scared of talking to one!  Both were very friendly to talk to, seemed younger in person than they do on TV, and not as tired as I imagined they might be.  I guess they are running on adrenaline.  

As for the debate content itself, that’s something you can make up your own mind on, having been able to see that from home.  

Thanks again to TVNZ for the invite to come along and watch!

In response to Peter Parussini of TVNZ’s comment on Kiwiblog:

“Throng chose to base their criticism on the proposition that there was something wrong and dodgy about using Reach as a measure.  That may be their genuine belief, but it is ill-founded and without justification.”

This is completely incorrect.

My criticism was that the measure being used (reach) was not clearly defined and as such, people were misled about what the figure being quoted actually meant. As I had said in emails to TVNZ that you would have been privy to:

“Just been looking through old press releases and the usual figure quoted is average audience and whenever the reach figure is used in some way, it is clearly explained as such so that people understand the difference in figures.  Statistics can be very misleading without clarification and I think in this instance, it is leading people to make an incorrect conclusion about the viewership.”

In a later email, I clearly explain I have no issue with using reach as a measure:

“I have no issue with the use of reach as a measure of performance.  It makes perfect sense for a special event such as the Royal wedding, sporting matches and finales of reality shows.  It also makes sense to use it as an overall season figure (hence I did not question the overall reach for the season). I don’t recall seeing reach being averaged across episodes in a press release before.

Using a statistical measure without clearly defining it can easily lead people to be misled, as happened in this instance, and others were questioning the figures before I did.  In the absence of a clear definition, people tend to use the figure they see being used most commonly.  The figure most commonly mentioned in press releases, and those published, not just on Throng but in magazines, newspapers, elsewhere online etc is the ratings, or average ratings.

I find it contradictory that you do not find the use of the figures to be at all misleading, but as a result of my post you are changing your policy to further clarify any statistics used in your press releases.”

Having lectured statistics at The University of Auckland for almost ten years, leading and co-writing a course which teaches students how to critique statistical reporting, the press release is an example to me of one of common problems in the communication of statistics: clearly defining measures being used.  

“And the replacement story continued the line that the Go Girls press release was misleading.  It was not.”

As a result of the press release, people were misled and this can be seen in the comments on the original press release.  Misleading means “tending to confuse” (http://www.thefreedictionary.com/misleading).

TVNZ contradicts themselves because in this email to me they write:

“”People” have not been misled by the press release – a handful may have been confused by the different but equally legitimate methods of measurement.”

“Misleading statistics” is a common term for when people are confused and/or end up arriving at a wrong conclusion.  Using the word “misleading” is not a judgment on the motives of those providing the statistics.  Misleading someone can be inadvertent and may very well be the case in this instance.   

“Nielsens even told Throng there was nothing wrong with it.”

Nielsens told us that the reach figure quoted was the correct number but said they do not comment on the use of ratings figures.  My blog post does not take issue with the number but the lack of clarity about the measure being quoted.  Perhaps the wording around that figure is very clear to those in the industry but not necessarily to the public.  

My blog post was intended to be an educational one which explains how people were confused, what the two figures were and their differences in definition.  As a result of my blog post, TVNZ told me they would:  “ask Publicity to ensure that in future all viewership figures are attributed to Nielsens and identified by their appropriate categories, as we do at the foot of News and Current Affairs releases.”

I’m pleased that happened as a result of my post since understanding statistical information is often not something that is easy and I’m hopeful that this means less of these types of misunderstandings occur as a result of the change in policy.  

As we said to TVNZ “The updated post educates people about the different measures in the hopes of clearing up the confusion people experienced when reading the press release.  Good statistics is all about the clear communication of data, including clearly defining measures used, even if the numbers themselves are accurate.”


Just a few days after TVNZ’s press release full of ratings blunders about TV3’s show 60 Minutes where they quote figures way lower than in reality, they’ve sent out another ratings press release which a number of people ended up being misled by.  

Using a statistical term such as “average” in that press release was ambiguous.  Many would take it to mean the most-quoted statistic of average audience, rather than what TVNZ had used and intended for people to understand: average reach

The two measures are very different and can lead one to very different conclusions about the popularity of a show.   

As a result, TVNZ has requested their publicity department ensure that in future all viewership figures are attributed to Nielsens and identified by their appropriate categories.

The two Go Girls figures, for comparison are:

Average reach: 708,200
Average audience: 348,000 (approximately half)

In case you’re wondering why the big difference in those figures, here’s what they’re measuring:

Cumulative Audience, also known as reach: 
Relates to the total number of different people within the selected demographic who tuned into the selected time period for 8 minutes or more (i.e. reached at least once by a specific schedule or advertisement). It is usually represented in thousands, but can be transferred into a percentage of the potential audience.

Average audience: 

The average number of people who tuned into the given time selected.

Reach is often used for special events such as the Royal wedding, sporting matches and finales of reality shows – where it makes sense to think about the total number of people who have watched some of the show.

Average audience is the most quoted figure, and the one most people would think of when it comes to TV viewership but of course reach gives you a bigger figure.