Statistics concepts and applications

Animal Testing Statistics

In research, reports, and activism efforts, statistics are often used to strengthen a specific cause or viewpoint. The challenge, particularly from the viewpoint of the reader, is that many of these statistics, while not inaccurately quoted, tend to be taken out of context. This creates an inaccurate focus that was unintended when the statistics were created in the first place. This phenomenon is clear in the guest post by Robin Lovell-Badge (2013) where the author makes a claim about the accuracy of quoting animal testing statistics to strengthen the cause against such testing on animals.

What particularly surprised me about the post is the explanation of the usefulness of animal testing in developing medicines. There appears to be a fundamental misunderstanding of the purpose of animal testing among animal activists who would have these terminated. While most drugs tested on animals fail to make it to the human trial phase, the purpose of such testing is not as much creating new drugs for human consumption as it focuses on protecting those people taking part in trials and protecting the public itself. In this context, the failure becomes useful rather than useless. Animal drug testing is useful for its ability to identify drugs that are potentially harmful or even fatal for human consumption.

Another thing that surprised me was the title. It made me question whether there are really that many statistics that are quoted out of context. Nine out of ten appears to be a very high number for misused and misquoted statistics. Of course, I can understand how activists and politicians would be tempted to use statistics for their own gain, but surely there are many, many research studies and other academic contexts within which statistics are not misrepresented.

I believe that many people are eager to believe reported statistics for several reasons. First, statistics tend to impart a sense of confidence in what is being said. Statistics about any subject provides a certain ideal of support for whatever opinion or claim is being made, especially if the statistics are quoted from credible sources. Furthermore, readers tend to be somewhat lazy when it comes to the closer investigation of apparently legitimate facts such as statistics. Hence, most readers would tend to take this information at face value. A third main reason is that those who quote the statistics do so with the express purpose of convincing readers or listeners to adopt or agree with their viewpoint. Many of the audience for such persons, however, are already of the same opinion as the authors using the statistics. Hence, statistics are seen as a factual validation of their viewpoint and they accept these at face value without further investigation.

Statistics are most often used out of context in politics. This is particularly the case close to election times, when all the candidates are desperate to gain the upper hand. The ruling party, for example, might make claims about what has been accomplished in terms of job creation, housing and welfare. The statistics quoted in such cases would perhaps be compared with statistics from a number of years ago in order to maximize the appearance of effectiveness. Job market statistics, for example, could be compared with those from five years ago, whereas comparison with three years or one year ago would tell a far different story and show the candidate in question to be far less efficient than the claim. Hence, statistics are not manufactured, but they are manipulated in order to benefit the election candidate in question.

Another platform from which statistics tend to be creatively decontextualized is advertising. An advertiser, much like a politician selling him- or herself, is concerned with selling a product. In order to accomplish this, the advertising must create a platform of desirability for the product. Statistics are often used to do this. For a beauty product, for example, statistics would be used to make claims about its effectiveness. Many of these ads make statements like "8 out of 10 women report significant reduction in wrinkles." While this is an impressive statistic, and is more than likely perfectly true, it fails to relate the context in terms of how many women have in fact been tested to arrive at the statistics. This type of context would provide a background in terms of the reliability and generalizability of the data being described. Statistics such as "nine out of ten" can be extremely misleading in this way, if not contextualized correctly.

In both the above cases, statistics are used somewhat out of context for the main purpose of creating a certain claim for the public to believe. Not contextualizing this information appropriately, whether intentionally or otherwise, creates a platform of deception and inaccuracy that the public is seldom aware of.