You Cannot Build A House On The Backs Of Straw Men

Written May. 5, 2010 by Tom Webster in Internet Radio + Terrestrial Radio with 2 Comments

We're very proud of the work we do here at Edison - not only for you, the radio industry, but also for our clients in politics, new/social media, consumer products and the many other industries we serve as a leader in opinion research. When you assume a leadership role, you take a few shots. It comes with the territory, and its generally unproductive to respond. In the odd case, however, these shots miss wide, and end up doing more collateral damage to our industry and our clients than they could ever do to us, and it is on those occasions that we are obligated to, at the very least, correct some inaccuracies.

In Mark Ramsey's post today, "My Internet Radio Stream Is Bigger Than Yours," we have one such shot, which appears to be directed at Edison, Arbitron, Jacobs Media, Inside Radio, Mary Beth Garber and the Southern California Broadcasters Association, and - ultimately - survey research, which is ironic, as Mark has at various times been a survey researcher. Do read Mark's piece - I encourage you to - but do also consider the flaws in his argument.

Mark's post was written as a refutation of this piece from Inside Radio:

Buzz aside, terrestrial webcast usage tops internet pureplays. A pair of recent studies shows that online radio listeners prefer pureplay internet radio stations over the streams of terrestrial stations. But unpublished results from Edison Research and Arbitron's 2010 Infinite Dial survey obtained by Inside Radio paint a different picture. The researchers found that 47% of Americans 12+ have listened to the stream of an AM/FM station, more than double the amount (22%) that have ever listened to "online audio from internet-only sources."
As one of the co-authors of this study, I can assure you that we stand behind those results. Mark's argument is a bit abstruse, but essentially he challenges this data by noting that the "100% accountable statistics published by Ando Media...routinely show that the usage of Pandora outstrips the usage of terrestrial radio streams, even when you aggregate some of the top radio groups together."

We are big fans of the work Ando Media does and are certainly friends with many from their parent company, Triton Digital Media (who, in the interest of proper disclosure, are clients of Mark but not Edison clients). Mark is 100% correct in noting that Pandora is the clear leader in streams today; however, this is so far afield of the Edison/Arbitron finding that it isn't merely apples-to-oranges, it's essentially a non-sequitur on at least three levels:

1. What the Infinite Dial survey shows is that a little more than twice as many Americans have ever listened to streams from terrestrial stations as have listened to online-only streams. This is a measure of people, not streams.

2. Furthermore, the question cherry-picked for this comparison refers to persons having ever listened to internet radio. Surely over the past decade this is true--Pandora's success as a mainstream play has been a fairly recent phenomenon. Mark is comparing a recent monthly ranker to what could be for some a 15+ year history of listening to radio online.

3. Finally, Mark states that Pandora outstrips usage of terrestrial radio even when you aggregate some of the top radio groups together, and later in the article notes that there are as many tuned to Pandora as CBS and Clear Channel combined. Again, this is apples-to-oranges to our question. We did not ask a representative sample of Americans if they had ever listened to the streams of "some of the top radio groups" or "CBS and Clear Channel combined," but any terrestrial radio stream. This is compounded by the fact that while Ando measures much of online radio, they don't measure all of it. Again, we are measuring Americans, while webcast metrics measure streams.

Most troubling, however, is the tacit assault on survey research in general, which is what truly prompted me to write this response. In the service of his argument, Mark positions survey research against server research, but this is simplistic and, at best, an intellectual shortcut in an argument populated largely by straw men. Mark notes that the "competing" webcast metrics "are not estimates. These are not polls. These are not opinion surveys. These are not samples. These are accurate accountings of all stream usage among the pure plays and broadcasters who allow their usage data to be measured by Ando."

Webcast metrics do indeed measure things that survey research is not well-equipped to measure. Certainly, I'd trust a server-based accounting of streams and sessions six-ways-to-Sunday before I'd buy a survey on the same data. Measuring unique Americans, however, and gaining insight into why those Americans use Internet radio, is the purview of survey research, and to pit the two against each other is truly a false choice - they are not only highly complementary, they need each other to truly provide the full picture on media consumption, advertising effectiveness and what really rings the till. I have to believe that Mark, who again provided survey research for clients in the past, believes this as strongly as we do. Perhaps, however, Mark's breathless pronouncement that " redundant to nothing and complimentary [sic] to everything" belies the fact that his criticism is less a polemic against survey research, and more of an emotional assertion that in his mind the battle for ears online is already over.

Indeed, Mark ends his piece on this note: "if you want to be assured the online radio sky isn't falling, I'm sure I can find a blue sky survey somewhere out there." We are not in the business of 'blue sky surveys.' On the topic of Pandora, in fact, I wrote this piece last week, and I hope you'll agree that I'm far from a Pollyanna on the topic. There are plenty of wake-up calls for the broadcast industry in our most recent Infinite Dial research report, and I invite you to spend some time with it and feel free to ask us your questions - the hard ones, and the easy ones. For our part, we pledge to do more than just critique; we will continue to create content, resources, and, yes, research to help you create your future. Ultimately, you will be the ones to make the hard choices, and to execute moving forward. Mark is right that the "whose stream is bigger" argument is frivolous - so why have it?

Reader Comments

Your 2¢, in chronological order — add your comment below.
1  Terry Purvis on May 13, 2010 2:25 AM

Mark has a habit of declaring the results of research as "facts", rather than "suggests". His ignorance of how server statistics are derived is rather naive to say the least. They aren't fact, but a thumb-suck, once you look into assumptions used. Sadly when you question him over it he gets defensive and rude. Ultimately this just shows his "advice" isn't worth the trouble of reading it. A rank amateur I'm afraid.

2  Tom Webster on May 13, 2010 8:21 AM

It's not my intent to spur an ad hominem attack on Mark personally, Terry--he's one of the smart people in the radio business. But both survey and server-based research provide estimates, as you suggest. They have differing strengths and weaknesses, and to position one as "fact" against the other as "guess" is, at best, disingenuous.

Add Your Comment

No <p> tags necessary, valid XHTML is always appreciated.

Edison Research

Receive new research and insight first. Subscribe to the Edison Research mailing list today!

First Name
Last Name
Email Address

What updates would you like to receive?

Election Research Updates
Broadcast Media Research Updates
Technology & Internet Research Updates
Consumer and Opinion Research Updates

Search The Infinite Dial

WWW Infinite Dial

About The Infinite Dial

No longer bound 'between 88 and 108 on your local FM Dial', radio has been liberated and now can be found virtually anywhere. This is a site to track radio in all its forms.

We are fans of great radio, whether it be on AM, FM, Satellite, Internet, HD, a Podcast, in any country on earth, or on any platform. The Infinite Dial will explore, analyze, and keep you informed about all the intersections of broadcast media and technology.

Have something to contribute? Just pop us a note and we'll get right back to you!