For Immediate Release:
Tuesday, March 9, 2004

Since its inception as an audience measurement company, Arbitron has made little change in its methodology and presentation of ratings results.   In the 1950's, when the service began, radio was very different from its current day descendant. In the 1950's radio programming was programmed in fifteen minute segments, each segment often completely different in content from the one preceding it. This was the basis for Arbitron's average quarter hour data in its reports. They wanted to accurately measure the audience flow from one program to the next.

In recent years, radio programmers have approached their programming hours much differently. With the advent of popular music in the early 60's, radio did away with fifteen-minute segments which varied one to the next and devised a much better way of holding audiences through each hour; the programming philosophy changed from a "segment" mentality to one that was more "cohesive" and the ancestor of modern day radio programming was born.

However, Arbitron did not change with the times. The company continued to measure radio in fifteen minute segments despite the fact that radio was no longer listened to specifically by quarter hour. Listeners began lengthening their listening events beyond fifteen minute intervals as the programming became more sophisticated. In order to maintain their methodology, Arbitron needed to adjust its reporting so it could measure audiences that continued listening beyond quarter hour benchmarks.   Thus began Arbitron's attempt to measure Time Spent Listening or how long listeners remained with a station beyond their first fifteen minute listening event. The idea of Time Spent Listening holds more promise than the reality.

Ask anyone who has visited Arbitron's facilities to review diaries and they will tell you that the process of measuring listening, especially time-spent-listening is pure fiction. A major research project in the 90's by an independent firm seeking to uncover the myths about diary-keepers found that most diary-keepers were vigilant about recording their listening on day one and day seven of their diary entries. All other information written in diaries was a result of attempts at several-day recall on the part of diary households. The study went on to reveal that in many cases only one member of the family would fill out diaries for all family members due to apathy on the part of the household. Those who took on the responsibility of filling out diaries did so from recall or reported recall and simply guessed or estimated the stations that were listened to and for how long.  

The reality of time spent listening cannot be accurately reflected with diary methodology because most diary entries are gross estimates of listening.

Cume listening is a much more reliable foundational statistic. Bridge Ratings reduces the amount of recall from seven days to 24 hours.   This shorter recall, combined with the more accurate telephone sampling methodology, shows increases of up to 75% in the cume listening reported in our survey.   Quite simply, the Bridge Ratings methodology is more effective at capturing a survey participant's actual radio listening.

From this very stable cume sample, Bridge Ratings asks "Of the stations you listened to this week, which is your favorite, the one you listen to most?"   Historical analysis shows that listeners contribute 66% - 85% of their weekly radio listening to the station they consider to be their favorite.

Therefore, the combined measurement of cume and favoriteness gives a station a much more reliable result upon which to base a) the health of the radio station and b) the size of the loyal audience advertisers seek when buying radio time. Advertisers know that by placing advertising on radio stations with a high (40%+) cume-to-favoriteness conversion, there is a higher likelihood of their message being heard frequently.

Average Quarter Hour is an artificial number.   It is generated by guesswork. It is also a combination of all variations of listening levels. It mixes heavy radio users with light users and creates a number which, on its surface, is difficult to value. Station A's average quarter hour of 20,000 persons is not the same as Station B's. As much as 60% of Station A's average quarter hour could be primary or heavy users. If Station B's 20,000 average quarter hour is composed of only 20% heavy users, which station will get better results for its clients and for its on-air promotion? The station with a higher degree of loyal heavy users will. This nugget is the statistic Bridge Ratings mines in our methodology that makes the difference.

Bridge Ratings measures listeners and their behavior more effectively than any other audience measurement service available today. We measure Cume and its performance twin, Favoriteness. Together they better represent a station's audience delivery.

End