PPM Explained

Math is hard, and lots of people are bad at it. They find it intimidating and frustrating. I include myself. It's a shame really since math is one of the few things which can contain actual, unadulterated, indisputable, non-partisan truth. To that point, despite those good qualities, it can be hard to understand. In that category PPM is so complicated that it's nearly impossible to explain to a layperson at all.

What's changed is how listening is tracked. So lets start there. In PPM a signal is embedded in the radio stations broadcast. But what kind of signal? How is it embedded? the signal is called a watermark. It is a sub-audible signal in the audio. It's not in the carrier wave, it's not RDS. It's actually in the audio. Arbitron actually has a device in the transmission chain that embeds a station identification code into the audio at each individual radio station.

Watermarking isn't exactly inaudible. I've done a few controlled tests to see if I can hear a supposedly inaudible watermark. Interestingly, I can detect it by ear, but only when comparing to source audio (about 70% of the time.) The watermark is a modification of the source audio that either adds to or changes existing audio. Watermarks are measured in three very specific ways. Robustness, Perceptibility and Capacity. These first two exist in total opposition. RBR reported that this had even come into play with PPM deployment. I quote:

"Anecdotal reports from radio broadcasters say that Arbitron lowered the watermarking energy in response to complaints about the watermarking being audible in certain circumstances."

A digital watermark is called robust with if it is detectable in the marked audio even if degraded by a number of transformations. Perceptibility is more binary. Since a watermark you can find by ear is useless we have a mushy adjective called "non-intrusive." Some watermarks are truly imperceptible, but the best ones are perceptible but non-intrusive. Capacity is how much data you can put in it.There are three primary ways to embed water marks. They can be used in tandem: Spread Spectrum, Amplitude Modulation, and Quantization. You probably figured out by the quote that PPM includes some Amplitude modulation. The PPM system takes 10 audio bands between 1 and 3 kHz. In the image above I've increased the amplitude of that region so you can see all ten peaks. But to reduce perceptibility, all 10 bands don't run full-tilt all the time. Each channel will have 4 frequencies within it. The identifying data will only be injected in one of them at any time, and only then if the energy (amplitude) of the channel is great enough to render the injected watermark imperceptible. This is of course patented.

Each of the four "injectable" frequencies represents 2 bits of data. I don't know the refresh rate but 500 ms rate was suggested and seems reasonable. This would produce a total of 40 bits per second or 2,400 bits per minute to encode a watermark. RBR suggested A unique ID might only consist of 50 bits. I think it could be done with less. There are only 14,000 stations in the US. Error rates will eat into this but multiple IDs per minute should be common. Compare that to a diary-based system.

PPM and College Radio

I was looking at the ratings for Boston this morning and I noticed something I didn't expect. WFNX only had a 0.8 share. They've been the big rock station in Boston for more than a decade and now they can't muster a 1-share. But the PPM books have had some outliers and odd results. Overall PPM seems to be more accurate. So when it's results don't jive with an assumption you have to reconsider your assumptions.So when I saw that WERS had a 0.9 Share... that was something that had to be reconciled. can a college station out-draw a commercial station? Hypothetically it's possible, it seems improbably when you think about marketing budgets but programming really is everything. In Boston WODS, an Oldies station is the number 1 station. That' s also unusual. So I took a look at the other PPM markets looks for this result. PPM is currently in the following 16 markets: Philadelphia, Houston, New York, Los Angeles, Chicago, San Francisco, Dallas, Atlanta, Washington DC, Detroit, Boston, Miami, Seattle, Phoenix, Minneapolis, and San Diego.

To preface this I'll say now that I excluded NPR talk and Classical as they obviously demo differently than rock. In New York, WSOU, WFUV, WFMU and WFDU don't even book WBGO came in at the bottom with a 0.3. That pattern continues almost without exception. In San Francisco KCSM gets a 0.5, in Dallas KEOM 0.6, in Houston KTSU 0.5, in Atlanta WCLK 0.5, in Philly WXPN has a 1.5, narrowly bested by WRTI with a 1.6, KPLU scored a 2.2 in Seattle, Minneapolis KCMP 1.6, and in San Diego KSDS got a 1.3.

Of that entire set of stations all are jazz except for three: Both KCMP and WXPN are AAA and KEOM runs a Hot AC mix of sorts. KEOM, while it's book was respectable was still dead last in it's market. WERS leans hard toward Triple A, and it's rating was comparable to the other two, this level of scrutiny proves little. The outlier could be instead a consistent and repeatable sampling problem. But, perhaps this is real and can set a new bar for non-commercial stations. Maybe they can compete on their own terms.

BREAKING NEWS

I avoid covering current events. I stand by that generalization. I always say that when I feel the need to cover them but of my average 250 some posts per year about two are about current events. Onward to the math...For months now we've all been reading about civic organizations, state attorney generals and radio stations suing Arbitron over PPM. I've written before about the natural predisposition of radio to reject all new numbers, especially any numbers that imply your listenership is down. But the reaction to PPM was way over the top.. I was even beginning to suspect that there was some validity to this just because of the wide-spread hostility. Then the story came out today. First thing today the Taylor on Radio-Info newsletter was in my inbox.

“Hispanic stations lost an average of 42% of their AQH ratings with 18-49s, urban stations an average of 34%…” That’s in the latest analysis across all 20 PPM markets done by consultant Randy Kabrich of Arbitron’s previous-diary-book ratings (not shares, but AQH ratings), compared to the latest June People Meter numbers. Kabrich says by comparison, “the non-ethnic-formatted stations lost an average of 16% of their ratings but held their ranks.” While the more serious loss of AQH ratings by the Hispanic and black stations had notable consequences for their ranks. Hispanic stations dropped an average of five ranks, compared to the diary. Urban stations dropped two ranks. Kabrich’s conclusion – “from diary to PPM, in currency ratings, the urban and Hispanic stations have more than double the loss of the non-ethnic stations.”

That just cant be right. First assumption. In the last 3 months 44% of Hispanic radio listeners didn't change the channel. Therefore the number was either too high before or too low. Around 4:30 FMQB sent out their email chiming in on the same topic with similar aplomb. they too were focused on Randy Kabrich, but hits a different more revealing quote:

"Share, which Arbitron uses, is not only not currency, it really can't be used to compare diary to PPM because of the vast differences in the PUR [Persons Using Radio]." Kabrich added, "Arbitron continues to use 12+ share in their comparisons which not only disguises a number of these issues, it is simply not currency and is irrelevant."

Randy is calling out Arbitron on the misuse of sample size. Sample size matters. But, here's why I think Randy is full of crap. Arbitron is guilty of using undersized samples. But, even previous to this the data was blatantly, criminally, unreliably, flawed! It was a diary system loaded with omission, falsehood, bias, non-response, error, and still yet some pretty small samples. If your data sucks it does not matter how big the sample is. It's just wrong. A bigger serving of crap is still crap. 60% more crap? Yes, still crap.

The PPM system removes every kind of error except the under-sized samples. If they samples are heinously undersized what we'll see is a wide variance between books. I don't see that. I think the shift is largely a correction. But not solely.

4 weeks ago Randy made a play for the same story but was focused on Urban radio. Story here. Edison makes the same assessment that I did: undersized samples vary, bias samples skew. Edison research makes the appropriate assessment of Randy's report:

"I simply couldn't understand what Randy was seeing. In the case of KMJQ (the blue line), one sees consistent upwards movement, while the PPM sample goes both up and down. KBXX (the red line) sort of stumbles along until their numbers shot up dramatically, again while the PPM sample gyrates around...We also ran an Analysis of Variance test (ANOVA) and found that the sample size is not associated with the AQH for either station."