Statistics are an important tool in analysis data, but often stats are misrepresented to give the impression the publisher wants to convey.
The recent “Media Buyers Survey” from Free TV Australia has some interesting stats. Based on an online survey of “around 100” media buyers from “Australia’s top media agencies” the survey is a measure of the perception the media buyers have of free to air TV, which is surprisingly similar to the information provided on the Free TV website.
The question is, does the Media Buyers Survey support the facts about the performance of Free TV or the marketing ability of Free TV Australia to influence the minds of their audience, the media buyers of Australia.
What do the advertisers themselves think? After all, the advertisers are the ones that are actually paying for the media, the media agencies are just advising and placing the media on their behalf.
I am not questioning the performance or otherwise of FTA TV, but I do object to the sloppy use of statistics to support a case, where the sample size and selection process is either misleading or flawed. There are standards for the design, analysis and presentation of survey statistics supported by the Australian Statistician.
On page 3 of the report they have a chart of “Household penetration of technology” that is referenced to their own estimate and various sources but no date. Then the balance of the report provides the response to very loaded questions like “Television is the best way to reach grocery buyers with children”. Was this the question to which they agreed or disagreed? Were they given a number of options, like agree, strongly agree, disagree, strongly disagree? Or was the question, please rank these media buy their ability to reach grocery buyers with children.
The results were mostly given as a percentage. But then only 86% had an opinion on page 9 as to if Free-to-air TV strengthens the performance of other media. Does this mean only 86% of people understood the question, or had an opinion, or did only 86 respondents answer it? We may never know because none of the axes were labelled.
Author: Darren Woolley