Radio rating is the percentage of total audience (or households) with their radio tuned to a particular station/frequency or program for a specified length of time over the total audience (or households) that have radio. As an important indicator of the attention economy, radio ratings are the currency of buying and selling of commercial airtime, provide an indication of radio program quality, and play a significant role in radio programs’ revisions, improvements and scheduling decisions. Radio rating is expressed as a simple percentage calculation, but it involves using comprehensive scientific methods in a series of processes that includes Establishment Survey, sampling, panel creation, data collection, data processing, etc.
-
Establishment Survey Establishment Survey, aimed at understanding the radio listenership universe and other detailed information of radio markets, is an essential part of the measurement service. Generally, an Establishment Survey is conducted by CSM Media Research at the initial stage of setting up a radio audience measurement system, after that it is conducted on a regular annual basis (every two years for some diary panels in small cities) to monitor changes in radio market over time and for projection of the measurement system's universe estimates. The Establishment Survey is the foundation of the RAM service, and has the three major objectives: 1) to provide universe estimates for the radio panel, including radio ownership, radio frequency reception capability, ownership of radio related equipment, and detailed demographic information; 2) to obtain targets for the results of the survey for various marginal and interlaced demographic and radio reception characteristics to which the RAM panel should conform; and 3) to provide a sample of households of known characteristics, to be used for RAM panel recruitment and maintenance.
-
Sampling and Setting up a Panel To give a reliable measurement service, a Panel must be representative of the population from which it is selected, apart from any deliberate disproportionate sampling. The achievement of a representative Panel is a two- stage process. First by employing a high quality probability sample design and then by ensuring the panel is, and remains, consistent with population values for a number of characteristics. Generally, CSM uses a stratified multi-stage PPS random sampling method to get the household samples and set up a panel. A panel data follows a given sample of individuals over time, and thus provides data in different time periods on each individual in the sample. Panel data possesses advantages over cross-sectional or time-series data by reducing considerably the omitted variable bias caused by sampling procedures, and offers statistical advantages. Radio audience measurement often employs panel survey, and the sample is relatively stable over time except a periodic turnover of a proportion of the panel. Continuous listening data on a 24-hours-a-day basis will be collected from panel homes.
-
Data Collection Basically, there are three types of data collection methods in radio audience measurement: telephone method, diary method, and PPM (Portable People Meter) method. The telephone method collects the listening data of radio audience using telephone survey techniques, and it uses single-round survey to gather information. Both diary method and PPM method use panel survey methods to collects the continuous listening data of radio audience. Diary Method The diary method uses diary cards to gather information on radio listening, and it requires all household members ages 10 or above in panel homes to record radio frequencies they listen to and the listening time throughout a normal day.Radio listeners can write down the code of radio channels they listened to, and mark the corresponding listening time band on CSM’s diary card. Each diary card can record one week’s (seven days) listening data for an individual. A time unit of 15 minutes is set for diaries, and if an individual has accumulatively listened to one frequency for more than 8 minutes during any particular 15 minute period, this frequency will be credited with 15 minutes of listening. Interviewers will collect the completed diary cards on a regular weekly basis. Virtual Meter Based on audio match theory, radio virtual meter rating system can match and identify the programmes audience are listening to and audio digital tag in audio library precisely and in real-time. Also, this system can monitor audience listening behaviors comprehensively at different time and locations with multiple-platform and multiple terminals, reproducing audience personal listening behaviors like online, offline, live, playback, interaction etc. measured in minutes; meanwhile, providing rating data.Virtual meter will be set up on smart phones that samples use every day, collecting and sending back audio information, and can measure listening behaviors of samples at different locations with different devices without changing listening habit or influencing listening behavior. This technology has been applied in business in the United States, Germany, South Korea, Poland etc. Continuous Radio Listening Survey In Continuous Radio Listening Survey, continuous listening data for radio channels will be collected from panel homes on the basis of 24 hours a day, 365 days a year.
-
Data Processing Data processing is an important part of the RAM service, and includes the following three major procedures: (1) Data entry and cleaning: raw data from the radio audience survey will be entered into the computer system (only for the diary method, and the PPM method can transfer the raw data to the computer system automatically). The data cleaning process will be conducted after the data are completely entered to thoroughly rule out any possible entry errors or logical errors to ensure data quality. (2) Data integration: After the data cleaning phase, comes the integration process. Listening data, demographic information, and radio program information will be integrated into a comprehensive Radio Audience Database. (3) Weighting and calculation: the weighting procedure will be applied to the survey data to restore it to agreed population projections. The weighting variables often include age, sex, etc. After the weighting, a variety of calculations will be performed on raw data from radio audience survey to produce different rating data.
