The pop-up survey has been one of the most positive contributions to web site research in the brief history of Internet research. The technique became widely used after the pioneering work of Micael Dahlen (1998) on the Swedish web site Passagen. So what is a pop-up survey and what are its benefits? A pop-up survey is a web-based questionnaire that appears in a new browser window as a person uses a web site. The survey itself is invoked by placing some JavaScript code in the web page. The same type of scripting is regularly seen on the web now to display in to interstitial adverts or to provide users of web sites with the additional information (e.g. pop-up help screens or search boxes). It has become popular for two main reasons. Firstly it is a proper sampling method (unlike self-selecting feedback boxes on web sites) and secondly because of the good response rates that can be obtained.
The Internet offers several versatile new technologies which can be applied not only to interviewing Internet users but also as a new and superior means to solve problems for which CAPl is presently used. Internet-based interviewing is not just for self-completion interviews but is useful to interviewers as well. Early work in this area has been concerned with how Internet interviews may be achieved at all, even with a simple questionnaire. This paper considers the next stage, when researchers will wish to present major questionnaires of the kind supported by other CAI tools. This paper examines the alternatives available for interviewing using the Internet, and considers the strengths and weaknesses of each in the context of major questionnaires. The conclusion of this analysis is that an organisation must be prepared to use different CAI techniques for different applications. The author presents a methodology for preparing surveys, based on this analysis, and shows its application in the case of the RegioLicht âElectronic Communityâ experiment in the Netherlands.
The term Asia is used here in a restricted sense to include the countries from India to the Pacific, bounded in the north by China, South Korea and Japan and in the south by Indonesia. This region contains approximately 2.7 billion people, about a half of the worldâs population. If any single factor could be said to describe this research environment, it is diversity. At one end of the spectrum are countries where market research has yet to gain any real foothold, among them being Myanmar (Burma), Laos and Cambodia. Some, such as Vietnam, are at the emergent stage. But most have research industries of several decadesâ standing, the full range of research services, and sampling and research methodology as sophisticated and variegated as western practice. The countries in this region certainly vary widely in the challenges they present for general population sampling, which are described below under five headings: 1. population characteristics; 2. information sources; 3. selection methods; 4. practical issues; 5. response rates.
This chapter sets out to explain the concepts behind the use of sampling in market research. It outlines some of the main options available to the survey designer and discusses the way sampling is carried out in practice. More complex topics are only introduced, with recommendations for more detailed reading.
In Canada, about 98% of all households have telephones, and unlisted telephone numbers are relatively rare. Given this, BUM has recruited participants for its radio ratings surveys using a sample frame of listed telephone numbers. In analyses of our typical sample profile, we have found that we tend to under-represent young, mobile, urban groups, the unemployed, students and those with lower income and education levels. We tested random digit dialing (RDD) in an attempt to improve representation of these groups. The differences between the RDD sample and a listed sample, however, seem to be quite small in the Canadian context. This paper compares an RDD sample with a listed sample, looking at demographic breakdowns, return rates, and tuning.
U.S. radio stations compete among themselves in a pattern that differs somewhat from todayâs population-proportionate survey samples. Some demographic groups are served by more stations than others, in ways that canât be explained by population size alone. We canât be sure whether the tendency of demographic groups to listen differently is the primary factor, or whether the patterns result from broadcasters choosing to pursue some demographics more vigorously than others. Because the demographic distribution of competition is a bit different than the demographic distribution of samples, the article closes with discussion of some challenges that might be involved with considering alternative sampling designs.
This paper deals with the measurement of audience reaction - what viewers think of the television programmes they watch - which is monitored on a continuous basis by broadcasters in the UK. The problems of translating appreciation into a simple score for each programme are discussed. The relationship between audience appreciation and audience size is examined and appreciation scores are found to differ for different types of programme and different demographic groups. Audience reactions to particular series are analysed and the way in which the service is being adapted to cope with the fragmentation in the television audience discussed together with the relevance of audience appreciation to advertisers.
From a panel organised by MEDIAMETRIE in January-February 1993 during three weeks based on a representative sample of French population, aged 15 and over (4,683 individuals), a treatment has been conducted in order to analyse answers from each participant by : - Duration of listening radio on three weeks - Share of listening dedicated to radio stations of Radio France network. Then, individuals who did not listened to Radio France have been classified into 7 different categories, in which listeners show an homogeneous listening behavior among them : - light radio listeners - light listeners to Radio France - average listeners to Radio France - heavy listeners to Radio France - strong radio listeners - very light listeners to Radio France - light listeners to Radio France - average listeners to Radio France - heavy listeners to Radio France For each level of listening, we describe : - audiences attitudes for different aggregate - socio-demographical profile of listeners. The survey's method by panel enabled us to evaluate public service radio reach. During the survey, half of the French population listened, at least once, one of the radio stations of Radio France network. It was interesting to define different ways of consuming these stations.
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.
Radio listening varies widely in terms of context and therefore, in quality and importance to the respondent and in accessibility to memory. On top of the core listening associated with regular behaviour choice and high motivation is a great deal of casual listening initiated by other family members or out of home. Techniques vary in their ability to capture the latter. The other main dimension of variation is the representativeness of the sample obtained; those very readily available tend to listen more. Generally, diary techniques yield higher levels of listening than recall and where studies are well conducted it is the higher levels that correspond most closely to validation by coincidental surveys. The evidence reviewed here identifies factors affecting response and shows that both techniques are very sensitive to variations in procedure. Generally it is important to make the diary as simple as possible and the station choices relevant and brief. For recall the more effort that is made to reconstruct the day's events the more listening is likely to be recalled. In Germany a thorough reconstruction of daily routines captured more radio listening than diaries. Sampling must be designed to capture the busier respondent and a correct demographic profile. Local broadcasting conditions are likely to interact with some of these findings and 'harmonisation' may not be easy. Possible electronic developments such as a personal meter identifying signal source are awaited with interest.
This paper looks at the results of several years of research delving into the elements that, in the eyes and ears of radio listeners, help make up their ideal 11 radio stations. The findings come as a result of a series of quantitative and qualitative studies designed to investigate the wants and needs of listeners and in turn, how radio stations can best go about meeting these wants and needs in an effort to build audience share. Special attention was paid to those programming elements which, when implemented as part of an overall strategic plan encompassing key marketing elements, assists stations in today's crowded radio environment to gain a competitive edge over others in the market. Specifically, the quantitative techniques involved using Quadrantâs Omnibus Studies combined with larger scale Listener Perceptual Studies, the latter of which using K-Means cluster analysis to assist in determining the appropriate segments of the market in which a particular client station should further focus their programming, marketing, and research thrust. Further quantitative techniques followed with the use of large scale Auditorium Music Studies, involving the testing of specific hook lines from music proposed to be played on the specific radio station. In each case a minimum sample of 120 respondents who fitted certain radio listening, target demographics and preferred music styles (as derived from the Listener Perceptual Studies) were recruited in order to ascertain those songs which were most popular and could therefore be played in higher rotation, versus those which were less popular and therefore should either be removed from the playlist consideration or played at lower levels of exposure on the station. Supplementing this major overview is ongoing research which is designed to update the information which has been gleaned over the past 4 years. This includes Quadrant's Compared to Five Years Ago, a quantitative study that examines radio listeners' responses to a series of programming and marketing questions with the objective to gauge how radio listening needs within demographics are/have changed over the past five years. This has assisted us, particularly in the areas of information provision, to tailor news services and community service announcements so that the information provided better suits the lifestyle needs of particular demographics, rather than trying to be a broad-based news and information supplier. In short, targeting news and information in a similar fashion to the way we go about targeting music styles to particular demographics.
In survey research it is very rare for all respondents in a given population to be interviewed. We usually take a sample of that population. The reason why we can do this is because a sample can give us, not necessarily the accuracy of a census (or full count), but sufficient accuracy for prediction purposes. This is true if the sample is representative of the population from which it is drawn. There are various sampling methods that can be used if we wish to obtain a representative sample. Such samples can give, depending mainly on the size of the sample, results to given levels of precision.