The application of cluster and biplot analyssis for television program rating data

Abstract

Television is one of the greatest inventions in 20th century. It can provide people with the actual news and information and also at the same time give them entertainment. The television industry has grown rapidly in the past 15 years since the government permits the establishment of commercial TV stations. As the medium of information, television has become the main choice for companies to advertise their products. Since there are eleven TV stations in Indonesia, advertisers have to make plan and strategy to broadcast the advertisement. Through proper statistical analysis, the information about which TV station should be chosen and when to broadcast the advertisement can be obtained. Data used in this study was secondary data from Nielsen Media Index from February 1-29th 2004, which was subscribed by Hotline Advertising. The data contains television rating for every programs that was broadcasted by eleven TV stations on February 2004. The results from cluster analysis on February 2004 categorize the eleventh TV stations into five groups/clusters. Each cluster has their own characteristics that are very useful for the advertisers. By constructing biplot, the information of strategic time on every day for each TV station can be obtained. The results from the biplot were suitable to explain the conditions that occur on February.