The video is unfortunately no longer available. Please read the text below, which describes the content of the video well.
Last Week Tonight is a television show hosted by John Oliver, a comedian who approaches complex topics related to world events, politics, healthcare, and social issues in a satirical and witty way. Broadcasted on premium television channels (e.g., HBO, The Comedy Channel, and Sky Comedy), the show is also available internationally on YouTube.
The episode from May 2016 addresses misinformation generated by scientific studies, fake news, and the media when it distorts information to promote sensational facts and gain public attention. The public is as such faced with two levels of distortion – first, the information and study results are affected by inadvertencies and bias in the research process and then by the lack of or insufficient peer review, predatory publishing, and selective presentation of results. In the next step, the media distorts the information further by altering the meaning of data, taking it out of context, and failing to include important details.
This can have disastrous consequences for the population (e.g., decreased level of well-being or misuse of over-the-counter supplements or medicines) and can also lead to diminishing trust in science, innovations, and future advancements.
00:22: Oliver addresses the growing number of television news programs and magazines that promote untrustworthy research results such as drinking a glass of red wine is equivalent to one hour of gym time or smelling farts might prevent cancer. And study results often contradict each other, which only adds to the confusion and misinformation.
02:31: For those who as a result tend to lose their trust in science and catalog it as bullshit, Oliver advocates for self-study and critical thinking. Stressing the idea that not all science is created equal, he identifies some reasons why low-quality studies still become public. Featuring Dr. Brian Nosek, a researcher within the Center for Open Science, he shows that standards of academic performance pressure scientists to regularly publish papers, even though they may not have discovered anything significant. To present interesting results with high frequency, scientists turn to various tricks such as altering the study duration, lowering the sample size, or p-hacking. Also referred to as data dredging or data snooping, p-hacking occurs when scientists keep combining the data until they obtain the desired result. Replication studies (i.e., other scientists conducting the same experiment) could be a solution to this problem. However, as Dr. Elizabeth Iorns from Science Exchange suggests, these kinds of studies get no visibility nor funds and are less valued than the original ones. Or as Oliver sums it up, There is no Nobel Prize for fact-checking.
05:09: While scientists might check the broader pool of evidence when confronted with research results, the general public takes the information at face value. Furthermore, in their quest for sensationalism, the media blows things out of proportion when they share the data with their lay audience. As an exemplification, Oliver introduces a study investigating the effects of high and low flavonoid chocolate during pregnancy, which concluded that there was no significant difference between them. However, the message was distorted when it was presented in a press release as The Benefits of Chocolate During Pregnancy.
07:40: Trying to satisfy the audience’s interest in fun, poppy science, media producers don’t always evaluate the scientific soundness of the studies, especially in terms of how representative the sample is. A widespread practice is to draw conclusions using a small sample that is not representative of the general population (e.g., a study that observed certain behaviors in 20 women cannot claim that this applies to all women).
09:02: In addition, when sharing the results with a broader audience, media representatives sometimes fail to mention that the study has not been conducted on humans but instead on rats. Oliver stresses that even if overwhelming evidence shows the treatments proved effective on lab mice, this does not mean that they will be successful in humans. Another pitfall consists of presenting research results too soon or without considering the complexity of the issue, as was the case in Paul Zak’s TED Talk about the "birth hormone O." that can positively change people in various ways.
12:01: Another source of misinformation is represented by industry-funded studies whose results support a particular organization’s goals or products. This was, for example, the case with a survey claiming that driving dehydrated is as dangerous as driving drunk. Apart from the fact that it was conducted on a sample of only 11 subjects, the study was funded by the European Hydration Institute, a foundation heavily sponsored by Coca-Cola.
13:28: Oliver then introduces a chart that presents contradictory study results showing that the same ingredient can either prevent or cause cancer and debates the idea of a la carte science. He also reveals the media’s tendency to cherry-pick the studies that best relate to particular convictions, which sometimes leads to extremism.
15:42: Under the name of TODD Talks (a.k.a. Trends, Observations, and Dangerous Drivel), Oliver satirically introduces a platform for simplified science that follows the structure of the morning news shows. Through witty comparisons, he stresses the idea that real science is a complex, slow, and rigorous process that does not easily lend itself to sweeping conclusions.
There are several reasons why people cannot evaluate misinformation in science, for example, a lack of knowledge regarding proper scientific methods and a lack of time and interest in applying these methods and checking multiple sources. Moreover, the inadequacies in how the scientific method is applied is an additional reason for the deterioration of public trust in science.