The authors report findings from an extensive analysis comparing two TV meter panels in Quebec and Canada; the TNS Picture Matching technology (PMT) and Arbitron's Portable People meter (PPM). Paramount in this comparison is the passive nature of reported viewing in the PPM panel versus the button-pushing task required of the PMT panel. Running in parallel for almost one year, the panels reveal important information on compliance in both the 'carrying' task and the button pushing task, individual viewing behaviour, the efficacy of the PPM technology in a complex TV measurement environment, and the reasons for differences in the audience measurements. The authors provide an assessment of degree to which various factors impact the ratings differences between button pushing and passive meter systems, and argue that button-pushing non-compliance accounts for most of the difference.
This paper describes a cross-sectional analysis of three people meter panels in Canada, one of which has been operating since 1997. With no formal 'forced turnover' policy, BBM Canada is ideally suited to examining 'time in panel' or 'panel fatigue' effects on a panel containing households of both long and short tenure. The cross-sectional analysis reveals no significant panel fatigue for adults, after controlling for other panel balance and control variables. Results for children, where fatigue is thought to be a bigger problem, are stronger but still not significant. Overall, interesting variations and consistencies across disaggregate models are produced including the positive relationship of the contact person, and the significant role of current panel control variables.
This paper provides a summary of BBM Canada's picture matching experience over the last four years. The authors focus on how well picture matching compares with DFM, how multimatching is handled, the problem of unmatched viewing, picture matching's ability to handle various delivery systems, and the extent to which picture matching can handle viewer behaviours like fast channel changing and channel surfing. The results of the assessment after four years indicate that picture matching is handling a very complex TV environment quite well. However, there are points for discussion including other uses for the TV that might need to be measured, the degree of granularity in measurement that is acceptable, and, of course, the issue of sample size.
This paper describes the Taylor Nelson AGB Picture Matching Technology (PMT) for identifying the program or channel being viewed on a television set. It outlines the structure of a field test conducted in Vancouver, Canada in late 1997/early 1998 to compare picture matching technology with the more conventional Frequency Measurement Technology (FMT). The results of the successful test are presented and discussed. The paper also describes and comments on the open and public character of the BBM Validation Test.
This paper is about ways to collect and integrate information on television/radio use on the one hand and use of different products on the other. And from a Canadian perspective; which is probably sufficient motivation for the reader to turn speedily to the next article. ("Whither product/media in Canada?"). "Product/Media" has been employed to describe two different types of analyses. The first uses in-home scanning data and television meter data from the same panel of households and attempts to discover relationships between advertising weight and product purchase (Jones, 1995). We are not talking about this. The second type, which is what we're talking about, collects information on media use as well as past, or generalized, use of products, services and stores, in order to help in media selection. These techniques help to describe the audience to television programs or radio stations, in terms of their use of various products or services.
Every researcher is asked the question: how big should the sample be? And every researcher has the same standard buck-passing answer: it depends on how accurately you want to measure what you measure. But given the amount of money that is traded on ratings numbers, it is important for the user to know what's real and what's statistical bounce in the surveys and therefore what the size of the sample should be to reduce this bounce to acceptable levels. The trouble is that the simple textbook formula we all know Vpq/n doesn't apply to the complex sample design and estimation procedures generally used in radio rating surveys. But techniques are available to estimate sampling errors empirically. BBM studies using such techniques show that there are two main influences on the size of sampling error and that they pull in different directions. The use of more than one respondent per household common in diary surveys tends to increase sampling error, and to increase it more, the wider the demographic. The use of average quarter-hour estimates tends to decrease the sampling error, and to decrease it more, the longer the time block being averaged. Generally, the latter effect dominates the former, meaning that the rating estimates are more reliable than the user might perhaps think. The details of sample design do matter too i. e. things like stratification, estimation procedures and particularly, the weighting scheme. We provide a case study of how attention to small technical details can pay off in increased precision just as much as an explicit increase in sample size: technique is as important as size.
There is increasing interest in the "Digital Superhighwayâ. How will TV audience measurement cope with this radical change in the medium? Many believe that the fast approaching fractionated and mobile television environment will necessitate a radically new measurement system that is at once portable, passive and people based. In an effort to find out what is going on in television audience measurement in other major countries around the world, BBM Bureau of Measurement commissioned New Electronic Media Science of New York to conduct a worldwide survey. The survey was designed to find out the methods currently in use for television audience measurement, levels of satisfaction with these present methods and the extent to which countries other than the U.S., Canada and Australia are moving towards passive people meters, portable or otherwise. This paper presents the results of the worldwide telephone survey. It appears that the push-button people meter is at the zenith of its development cycle, with systems operating in 30 of the 34 countries examined. Most of the systems have been introduced in the last three years, but a third of the people meter countries are already considering a shift to passive measurement -- a move that has strong support in almost all the countries surveyed, especially those with more experience of the conventional push button people meter.
Canada is a country of immigrants, and official government policy encourages bilingualism (English and French) and multiculturalism. Television services are provided in French and English across the country; cable and traditional over-the-air television stations offer programming in other languages. This presents many challenges for television audience measurement. This paper reviews the methodologies used in three recent studies designed to measure the television viewing of specific linguistic minorities, contrasting them with the procedures normally used for the regular television surveys of the general population conducted by BBM. It then reviews the results of the special studies, using them as benchmarks against which to measure the ability of the regular BBM surveys to measure ethnic minorities. Three questions are addressed: 1. Does the normal BBM survey process recruit the right proportion of members of linguistic minorities? 2. Is the sub-sample of the minority that is normally recruited representative of its population? 3. Does any of this make a difference to the published ratings results for the total population?
The self completed television viewing diary continues to be an important instrument for audience measurement in Canada and the U.S. The diary collects information on a 15 minute basis, assuming that only one channel is watched for all or most of the period. With increasing channel choice and the universality of remote controls this assumption is being challenged. Viewers change channels so frequently, it is alleged, that they cannot possibly be expected to record what they view. This paper addresses the allegation by analysing the extent and nature of actual channel changing behaviour. It uses data from a data base of channel changes, compiled from the BBM-Videoway television meter project in Toronto in Fall 1990. It shows that the extent and influence of "zapping" or "grazing" has been greatly exaggerated. Contrary to conventional wisdom, most sets stay tuned to one channel during a quarter hour period. Those tuned to more than one, stay with a principal channel during the great majority of the quarter hour, sampling another channel or two for just short periods of time. For most viewers there does appear to be one identifiable channel tuned for most of the quarter hour, as assumed by the paper diary. This is shown to be the case during different day parts and for different types of programming