Author: Gorm Palmgren

Fake news becomes dead serious for human health when the English doctor Andrew Wakefield publishes his fake research results in 1998. But many others are to blame in a new era of science scepticism where real experts are questioned and anyone can rise to experts.

By Gorm Palmgren, freelance science journalist writing for EUSJA’s NUCLEUS project

Had measles virus had hands, they would have rub them together in anticipation on Saturday, February 28, 1998. After being held brutally down by the effective MMR vaccine for decades, the door to freedom is now starting to open so they can finally get loose. The door to freedom is a scientific article published by the English doctor Andrew Wakefield of the Royal Free Hospital in London in the acknowledged journal The Lancet. In a study of 12 children, the doctor points to a correlation between autism, bowel disease and the MMR vaccine against measles, mumps and rubella. Wakefield has a theory that the combined vaccine attacks the intestines and somehow affects the brain, so the children develop autism within a few days.

The article stirs attention far beyond the ordinary research community, and already a few years later the consequences can be read directly in disease statistics. In large numbers, frightened parents choose not to give their children the MMR vaccine, and the measles virus is quick to exploit the vulnerability of the unprotected children. In just five years, the number of measles cases in Britain are doubled and most unfortunately, the parents and their children have received nothing in return from this high price. It turns out that Wakefield’s research is a scam, but this is not fully revealed until several years later.

Lawyers get paid to miscredit the vaccine

Andrew Wakefield is the son of two British doctors, and at the age of 24, in 1981, he graduates in medicine at St Mary’s Hospital Medical School in London, which at the time is Britain’s most popular medical university. He trains himself to become a surgeon specializing in organ transplants, but gradually he changes interests, and in 1993 he receives attention with a scientific article. In the article, the young doctor presents a theory that measles virus can cause Crohn’s disease. Two years later, he suggests that it is the living, but attenuated virus in the MMR vaccine, which is the cause of the disease, and soon he broadens the accusation and suggests that the vaccine may also cause autism.

Wakefield has reached his theories through studies of children, and gradually it appears that he has recruited his subjects in a highly unconventional way. When his son celebrates his 10th birthday in 1998, the doctor offers all the young guests £5 to donate a blood test to his experiments. During a lecture, the doctor later tells that two of the children fainted when the blood samples were taken while a third threw up in his mother’s lab. Such an approach is violating the ethical rules for medical trials, but it soon appears that Wakefield has made far more serious mistakes in his selection of subjects for his studies. This is revealed by British journalist Brian Deer when he digs into the case and in 2004 begins to present his disclosures in a series of articles in the Sunday Times and the medical journal BMJ.

The journalist can document that the parents of most of the subjects of the trial was from the beginning convinced that the MMR vaccine was responsible for their children’s autism. The parents are in the process of preparing a lawsuit against the pharmaceutical company behind the vaccine and, in anticipation of a generous fee, their lawyers pay Wakefield £435,643 to document the children’s symptoms. The journalist also reveals that the research results and the connection between the vaccine and autism are largely fraudulent. Three of the children who allegedly developed autism have never been diagnosed with the disease, and five of them already had mental and physical developmental disorders before being vaccinated.

Measles virus exploit the parents’ concerns

Immediately after the fake and disturbing results were announced, the UK doctor gives a press conference calling on parents not to vaccinate their children before researchers have further investigated the case. The following day, the story is all over media: “MMR vaccine causes autism.” Newspapers, tv-stations, radio and the net went mad, and like a snowball, the story grows bigger and bigger. It doesn’t culminate before 2002 – four years later – where British newspapers print 1531 articles on the subject, two thirds of which are critical of the MMR vaccine or directly refer to it as uncertain.

The negative press sends shock waves among parents, and from fear that their children will get autism they turn their back to the vaccine in large numbers. Before Wakefield published his results in 1998, 90 per cent of British children received the vaccine, but in 2003 that number was reduced to 80 per cent, and a similar trend can be seen in most other countries. The drop may not sound critical, but it is. Vaccines are only effective if the pathogenic bacterium or virus doesn’t have a safe reservoir in the form of unvaccinated persons whom it may infect and propagate in.

WHO recommends that at least 95% of the population is vaccinated against measles. In that case, measles virus has so bad living conditions in the population as a whole that everyone – even those who are not vaccinated – receives protection. Accordingly, it is a collective responsibility to be vaccinated, and by opting out, you expose not only yourself or your children but also put all your other family members, friends, colleagues and fellow citizens at risk of serious illness. Finland long ago achieved the target of 95%, which has led to the eradication of measles which has not showed up in the country for the past twenty years. Britain was also well on its way reach the goal, but was set back by Wakefield’s scam, which instead launched a new unfortunate development.

Measles virus doesn’t hesitate to exploit the open door to freedom. After the number of measles cases in the UK has remained relatively constant at 100-200 per year through the 1990s, the number of disease cases now begins to rise. In 2003 it has more than doubled, and from then on it continues to escalate up and down with a steadily rising trend. In 2008, it reaches almost 1400, and in 2012, it sets a new record with over 2,000 Brits who get sick of measles. For the first time in 14 years, people start to die from acute measles infections and since 2006 the death toll amounts to four British children. It is not only in Britain that Wakefield’s trial horrifies parents in great numbers. Also in the rest of Europe, the proportion of vaccinated children drops and the number of measles are quadrupled from 2008-2011. The same is true in the United States, where less children receive the MMR vaccine than in countries such as Libya and Zimbabwe.

Hand-picked subjects give predictable results

The parents’ concerns, however, are grossly exaggerated. Right from the beginning, health authorities all over the world assure that the vaccine is both safe and useful, and they have historical evidence for that claim. In the United States, about 400,000 cases of measles were reported annually until 1963, when the measles vaccine was introduced to children around the age of one year (the combined MMR vaccine came in 1971). Just five years later, the number of cases fell to about 50,000 a year, and since 1990, when the vaccination program expanded to include an additional shot at the age of 4, there have been less than 200 cases of the dangerous disease annually.

No matter if Andrew Wakefield’s results had been fraudulent or not, the media should have been extremely wary of spreading the worrying message from the very beginning. It is fundamental in the scientific method that one should never feel too convinced of the results of a single experiment, even though it is published in a recognized journal such as The Lancet. Experiments should always be repeated several times in different ways in other laboratories, because biological systems are so complex that there can be many explanations for the apparent findings. Had the media been more critical in their review of Wakefield’s investigation, they should be able to notice that his conclusions were very weak. But the British doctor also violated several other basic principles in the scientific method, that a highly qualified science journalist should have noticed.

Firstly, his subjects were not randomly selected but carefully handpicked among a group of people who already suspect the MMR vaccine to be the cause of their children’s ailments. This makes the investigation self-confirming and thus very difficult to conclude anything from. Although one can easily find 100 healthy people who have smoked like chimneys throughout their lives, it’s obviously misleading to use it as an argument that smoking isn’t harmful to the health. Instead, you must randomly select 100 smokers and 100 non-smokers, after which you compare their health – and then you will undoubtedly come to a completely different conclusion. But Wakefield’s study was also weak because he used only 12 subjects and had no control in the experiment. In other words, he should have included children who had not been vaccinated or who did not develop autism.

After the Lancet paper was published in 1998, several other researchers start to search for a connection between the MMR vaccine and autism. Already in 2002, Kreesten Meldgaard Madsen of Aarhus University in Denmark publishes a comprehensive study covering all children born in Denmark from 1991-1998. Among the 537,303 children, 82 percent had received the MMR vaccine, while 738 had developed autism. Together with his colleagues, the researcher now uses statistical methods to see if there is a connection, and the short answer is: No. In fact, researchers find that the risk of autism is 8-17 per cent lower for the vaccinated children compared to those who have not been vaccinated. Intuitively, one would think that the MMR vaccine might then protect children against autism, but researchers estimate that the statistical uncertainty is too big to make such a conclusion.

Guy Eslick from the University of Sydney in Australia also acquits the vaccine on all charges in 2014. In a so-called meta-analysis, he has analysed five major studies, where other researchers have investigated the connection between autism and vaccines – including Kreesten Madsen’s study. Together, the studies include 1,256,407 children, and again the conclusion is clear: Children do not risk autism by being vaccinated. Against this conclusive result stands Andrew Wakefield’s study of 12 children – 100,000 times less than the Australian meta-analysis.

The test results are withdrawn

Andrew Wakefield does not have much confidence in the academic world, but what has been revealed of his scam so far is still only the top of the iceberg. The journalist Brian Deer reveals in 2004, that the doctor has previously applied for a patent for an alternative measles vaccine. The existing MMR vaccine is a mixture of three vaccines against measles, mumps and rubella, and according to Wakefield, the combined vaccine is harmful in this mixture. In his patent application, he explains that he will be able to avoid the risk of autism and bowel disease by giving the measles vaccine separately, and it is precisely such a vaccine he has applied for a patent. Wakefield, accordingly, has a clear financial interest in discrediting the existing MMR vaccine because that would pave the way for his own alternative and it provokes an unpleasant suspicion that the doctor deliberately cheated with the article in The Lancet for his own financial gain. The new accusations make ten of the twelve other researchers who co-authorised the contested article turn their back to it and regret the consequences of it.

Wakefield is now deeply discredited, and The Lancet is under pressure to retract the 1998 article. So far, it has only regretted certain irregularities in connection with the attorneys’ funding of the investigation, and the editors’ motive is likely that a withdrawal may put the highly acclaimed journal in a bad light. However, in January 2010, the UK General Medical Council (GMC) states that Wakefield has committed both scientific malpractice and unethical treatment of ill children by taking blood samples from them without prior approval. The same year, The Lancet finally retracts the fraudulent article and shortly after, the British health authorities withdraw Wakefield’s medical authorization. Slowly, worried parents around the world become more calm. More and more children get the MMR vaccine and in many countries, the target of 95% vaccinated children is within reach.

The HPV vaccine is the new bogeyman

Even though the worst storm is now cleared, several vaccine sceptics are working hard to cast doubts over vaccines. They are especially going after the HPV vaccine, introduced in several countries about 10 years ago.
This vaccine protects against the most common varieties of human papillomavirus (HPV) that mainly infect young women’s genitals and cause genital warts. But in addition, infection with HPV virus is responsible for virtually all cases of cervical and rectal cancer, as well as over half of all cases of head and neck cancer as well as cancers of the vagina, lips and penis. Worldwide, HPV virus is believed to be behind 300,000 cancer deaths every year, equivalent to 5 percent of all cancer victims.

The HPV vaccine has only been in use for about 10 years, and since it protects against cancer, which typically takes 10-30 years to develop after an HPV infection, it is still not known how effective it is. The most comprehensive study of effectiveness dates from 2017, where 2084 women from the Nordic countries were followed for 12 years after they had been vaccinated. None of the women developed neither cervical cancer nor the progenitor of the disease. Based on this study and more extensive clinical trials before the vaccine was approved and marketed, researchers estimate that it can prevent 70-90% of all HPV-related cancer cases. This means that the vaccine can save more than 200,000 lives a year.

Despite the promising prospects, the HPV vaccine – most commonly known under the Gardasil brand – has received many opponents who encourage parents not to vaccinate their children. In particular, they use blogs, social media and other online services to spread information that is at best misleading and, at worst, can be termed fake news. The Canadian anthropologist and neuropsychiatrist Anna Kata of McMaster University in Ontario, has studied these groups, which, in her opinion, have greater and greater influence on the parents’ decision to say yes or no to vaccines. There is a general tendency for people to listen less to doctors and other experts and instead to be guided by information that they find themselves on the internet.

The arguments of vaccine sceptics are weak and misleading

Anna Kata has found that anti-vaccine campaigns create widespread distrust of the experts. They use arguments like “scientists are bought by pharmaceutical companies” and “science has previously been wrong”. On the other hand, the simple gut feeling of common people are praised as credible with arguments like “I am the greatest expert in my own child” and “So many people [those who do not allow their children to be vaccinated] cannot be wrong”. Other arguments put the vaccine in a bad light by emphasizing that it is not 100% effective and therefore failing, or that one does not yet know exactly how effective it is, which makes it uncertain. And again, other arguments strongly condemn the vaccine to be unnatural as opposed to immunization after a natural infection, or potentially dangerous because it is impossible to prove the opposite.

In other words, the arguments of the vaccines sceptics are very emotional and somehow difficult to argue against because they are so vague and broadly formulated, often based on individual cases, and do not at all consider scientific documentation. Not even one of the most scientifically based arguments of the anti-vaccine movement can pass the test, namely that several young women have experienced chronic fatigue after being vaccinated. These unfortunate cases have, however, been thoroughly studied by researchers who, overall, have not been able to connect them with the HPV vaccine. A

study by the British health authorities from 2013 thus found that 29 vaccinated girls experienced chronic fatigue symptoms between 2008 and 2010, in which period more than 1.5 million girls received the vaccine. Even assuming that only 10% of all cases of chronic fatigue were reported and thus contributed to the researchers’ study, the number is still only within the limits of what would be expected because the symptoms can also arise from completely different and often unknown reasons. In addition, the researchers found that the number of cases of chronic fatigue in young girls did not increase after the vaccine was introduced in Britain in 2008, but actually decreased by about 10%.

Two other comprehensive studies from 2017 by Anders Hviid of the Danish research center Statens Serum Institute also show that the HPV vaccine also has no such long-term side effects. The two studies comprise over 5 million women who were either vaccinated or not and the researchers have followed them for several years. The results show that the vaccine does not increase the risk of serious autoimmune or neurological diseases, and that women do not have to fear for side effects on their children if they become pregnant at a later date.

Fraudulent doctor goes to the movie

Far away from the broken reputation in his native country, Andrew Wakefield has now settled on the other side of the Atlantic. In January 2013, he appeared in Washington DC for the Realscreen Summit, an annual recurring meeting where the television industry is trying to develop new concepts for reality shows. From his laptop, Wakefield showed a short trailer promoting his idea for a new television show called The Autism Team: Changing Lives. The trailer showed three children with autism who screamed wildly, bit their mother in her hand, and hit themselves in the head with a book. But to the relief of the unhappy parents, Wakefield and his autism team had developed an effective cure, and in the new reality show, one would follow the dramatic journey of the families from the suffering of the disease to the blessing of healing. So far, the TV show has not been aired, but the former doctor continues his efforts to blame vaccines for causing autism. In 2016, he celebrated the premiere of his documentary Vaxxed, claiming to reveal the authorities’ attempt to destroy data that allegedly showed a link between the MMR vaccine and autism.

Even though Andrew Wakefield is an expert in such scams himself, the movie received a cool reception and has not attracted nearly as much attention as his first attempt to spread fake news in 1998. And fortunately, the documentary has not triggered a new wave of mistrust among parents. Hopefully, they will let science rule and give their children the vaccines that can keep them safe from lethal infections with measles virus and HPV.

The ice cap is expanding on Antarctica, so there is nothing to be afraid of, and global warming is anyway entirely natural and for our own benefit. The claims of the climate sceptics are repeatedly rejected by researchers as either directly wrong or simply misleading. But the advocates of the man-made climate change are opposed to a well-orchestrated system that spreads dubious claims strategically and repeatedly, so you might surrender and start believing them in the end.

By Gorm Palmgren, freelance science journalist writing for EUSJA’s NUCLEUS project

On Sunday, July 9, 2017, the American alt-right news site Breibart News published an article on their website under the heading: “Almost all the newer global warming is being fabricated, a study shows.” The article referred to a so-called peer-reviewed scientific study, meaning that other scientists have scrutinized the scientific methods and conclusions in a fully formalized manner and finally assessed the results as credible. The article tells us how scientist Craig Idso and his two colleagues behind the study have analyzed the past 120 years of climate data and now can document that organisations such as NASA, as well as the United States and Britain meteorological institutes have consistently manipulated and misinterpreted the numbers. Accordingly, there is no scientific evidence of the allegations that global warming is taking place or that the recent years have been the warmest in Earth’s history.

In a few days, the article was shared over 30,000 times on facebook, it received almost 2,000 comments on Breitbart’s own website, and the news was followed up and discussed in many other media. The message ‘Global warming is fake news’ had now spread across the United States and the rest of the world, but something was entirely wrong. It soon turned out that both Breitbart’s article and the actual scientific study was wildly misleading and manipulating. It was in fact fake news.

Unfounded accusations on climate date corrections

This was revealed five days later by one of the United States many fact-checking organisations, Snopes.com, which tries to keep a watchful eye on what’s right and wrong in the public debate. Snopes.com found several flaws. First, Idso’s investigation is not, as reported by Breitbart, published in a peer-reviewed scientific journal, but rather only in one of the researchers’ own blog. And the so-called peer review covers that seven other researchers – including an economist, a retired space engineer and the leader of the Christian organisation ITEST, which equates science and religion – have declared their agreement in the conclusions. It has nothing to do with peer review, so the claim to legitimize the study as credible is false.

The study’s accusation that the historical climate data has been deliberately manipulated also proves to be highly misleading. Over the 120-year period, the established climate researchers have properly adjusted the collected data, but it is done, for example, to compare temperature measurements previously made with a mercury thermometer in the evening, but now measured with a digital thermometer in the morning. It is such absolutely necessary corrections, which are referred to as manipulation, and Idso is not pointing out why the corrections have been made. Breitbart’s review of the study also claims that the corrections consistently make the most recent temperature measurements higher, while the older ones are getting lower, which overall gives an enhanced impression of global warming.

The assertion is based on a graph that is shown in the article, but which actually shows something completely different than the journalist apparently believes. The graph does not show, as stated in Breitbart’s article, the difference between the original and corrected measurements, but on the other hand, it shows the differences between two sets of corrected measurements. Comparing the original and corrected measurements reveals out that the corrections actually make the old measurements before 1940 a little higher, while they do not change significantly on the measurements in the last 70 years. In this way, the corrections actually make the global warming look smaller, which is directly against Breitbart’s accusation, and thus cancelling out the entire basis for their claim that climate scientists manipulate data.

Popular mantras backfire

Common to many of the fake news abounding on the internet is that they often take off from a single detail that might be correct, but which is taken out of its proper context and blown up so that the overall picture gets distorted. The practice applies to both sides of the debate and can be exemplified by the well-known claim that “97 per cent of researchers agree that global warming is man-made.” The problem with the assertion is that nobody really knows where it comes from, who those referred to as climate scientists actually are, and what they really agree on. The claim does not refer to a particular study, which can be subjected to a critical review, but rather to a kind of average of a large number of unspecified studies on various scientist’s opinion about climate issues.

Some of them have exclusively questioned dedicated climate experts in meteorology, while others have included, for example, general meteorologists, geologists, physicists and hydrologists. And they each have operated with their own limits for when the researchers agreed in the statement – e.g. wether mankind is responsible for all global warming or just half of it. On such an undefined and vague basis, it doesn’t make sense to talk about 97 per cent consensus, and the climate change defenders – those who believe people have a responsibility for global warming – are thus nourishing an undocumented assertion in their efforts to win support for their cause. Such weak arguments, however, can easily cause backslash because they are easy to peel apart by climate change deniers.

To that end, the climate sceptical website Natural News in 2016 took a closer look at one of the above mentioned investigations that dated back to 2008 and argued that it was the original source of ‘the 97 per cent claim’. In the study, Maggie Zimmerman of the University of Illinois in Chicago found that 82 per cent of 3,146 climate-related researchers subscribed to the notion that humans carry a significant share of responsibility for climate change. But in her conclusion, she chose to ignore researchers in fields such as meteorology, geology and hydrology. Instead, she only took into account a very small group of 79 – out of the originally 3,146 respondents – who represented the most ‘specialized, active and knowledgeable’ climate scientists. With that small trick of data filtering, she managed to get over the magic border and land on 97.4 per cent support. Natural News argued – with good right – that this was gross manipulation, and in an article on their website, they used the study as proof that ‘the 97 per cent claim’ is fake news. The article was viewed 58,000 times by readers who were met with the captivating headline: “The REAL FALSE NEWS revealed: ’97 per cent of researchers agree on climate change ‘is a fabricated fraud number”.

Political messages are sold as advertisements

Headlines like that say the whole thing, so if you are already sceptical about the man-made climate problems, you are immediately confirmed in your opinion, whether or not you read the article. If you then choose to share the article on facebook or twitter, it begins a journey into cyperspace, which in many cases is anything but random. If the news had referred to an ordinary product such as diapers, the manufacturer would make sure that it was promoted as an advertisement. This will make it appear on more peoples feed and make it more visible. But it will not appear on just anybody’s feed but instead on the feed of carefully selected individuals.

Facebook collects a wealth of information about its users and advertisers can utilize that information by asking the social media to preferentially display the post to a carefully selected group that might be interested in the product. A diaper manufacturer can e.g. promote an add to young women with small children living in countries, where the company sells its products. But it can also further narrow in on women with a relatively high income and preference for organic products because the diapers are organic and therefore a little more expensive than other diapers. This type of targeted marketing is used all the time on social media, and that’s why you typically get ads for things you really care about and that you’ve just seen online. But targeted marketing is also widely used for political messages on facebook.

During the election campaign of the US presidential election in 2016, the Trump campaign used the majority of the $90 million digital advertising budget to display a daily average of 50,000 different variants of facebook ads tailored to individual users. On facebook, it’s easy to blur your identity so the sender of the advertisement isn’t always obvious. In September 2017, facebook announced that a Russian company was behind 3,000 political advertisements in the US election campaign worth $ 100,000 and seen 10 million times. A climate critical think tank may use exactly the same approach to promote a post that reveals man-made climate change as a scam. They can target people with a specific political affiliation, education level, job situation etc, and thus increase the likelihood that it will be well received and get shared a lot.

Electronic snitchers pursues you online

Fake news can therefore be sold and marketed as any other product. Like any other marketing, it’s not free, but costs around $ 1 each time a facebook user engages in the post by e.g. liking, sharing or clicking a link to the article. Therefore, organisations or companies with pockets full of money are always behind when false news is spread on the web. And when the topic is man-made climate change, the oil industry is very often involved. One of the most cited and prominent climate sceptics, Pat Michaels from the American think tank Cato Institute, acknowledged in CNN on 2010 that he gets about 40 percent of all of his funding from ExxonMobil.

ExxonMobil is the world’s largest oil company and has for decades had many climate sceptics other than Pat Michaels on their payroll list. Justin Farrell of Yale University in the United States surveyed the entire climate sceptical movement in 2015 and found that it was composed of 4,556 people with connections to 164 organisations. Together with a fund associated with another large oil company, Koch family foundations, ExxonMobil supports about half of these organisations, and specifically this half is considered to be the most active and influential. Another study has shown that among the ten most influential climate sceptics, all are affiliated with organisations funded by ExxonMobil.

Facebook, of course, keeps an eye on what messages you read, like and share on the social network, and with the help of so-called cookies they also keep an eye on what other websites you visit online. Cookies are a kind of electronic snitch, and if a website collaborates with facebook – for example, have ads with them or have a ‘like-button’ on their website – they are used to tell the social media that you have visited the site. This way, facebook and other social media as well as their political advertisers can lure your attitude to topics of the climate debate that you are presented with during the day.

One might imagine, for example, that a colleague during the morning coffee expresses his scepticism regarding a new report on alarming climate change. You get interested and find the report online, and at that very moment facebook register you as interested in the climate debate. To follow up on your colleague’s scepticism, you visit the critical website he has mentioned, and the social media registers you as a potential climate change denier. It will therefore target climate sceptical lookup to you, and during the day you will see more of that type on your feed. Gradually, it can give you the feeling that climate sceptics are all over and that there are many good reasons to doubt whether global warming is due to human CO2 emissions from fossil fuels.

If you intend to spread fake news, it’s about getting your story out to as many as possible in a short period of time. Several researchers have used mathematical models to calculate what is needed for an article to go viral on social media. First of all, it must get peoples immediate attention with either a provocative headline or a brilliant picture, so people will find it in their feed. But even if users share it, it will almost always die out quickly and not get shared a lot. Researchers such as Filippo Menczer of Indiana University in the United States have shown that if the probability of a post being shared is 1:10, then there is a probability of only 1:40 that one of those who received the shared post will share it themselves. And for being shared at the third level, the probability is only 1: 360. In order to go viral, something very special is needed and that is a so-called super sharer.

Bots are the new super sharers of the net

A super sharer is one who, contrary most social media users, doesn’t count its followers in hundreds but in tens of thousands. If a fake climate news story ends up in a super sharer’s feed and she chooses to share it, it will be distributed to a large number of​people and it will have a good opportunity to go viral. But precisely because the super sharer has so many connections, she also receives huge numbers of posts herself. Accordingly, there is only a small chance that she will spot the fake climate news in the first place and subsequently share it with her many followers. If you are a big interest organisation with a lot of fake news that you want to spread online, you need a super spreader who is always aware of these news. This can easily be obtained in the form of a bot, i.e. a piece of software or an algorithm that automatically writes, shares and comments on post on social media.

Bots are commonplace on twitter, where they are often used for mass distrubs of useful information such as weather alerts, and it is estimated that about 20 percent of all tweets are created and shared by bots. But they can also be used in more crafty ways. This was demonstrated by researchers at the University of Southern California who showed how so-called twitterbots participated in the debate up to the US presidential election in 2016. The researchers used the BotOrNot computer program to analyse 20 million tweets containing hashtags related to the election. The computer program uses artificial intelligence to assess whether a human being or a bot is behind the tweet. For example, by studying the wording in a tweet, consider when the tweet was made (a tweet sent in the middle of the night is more likely to be sent from a bot than a human being) or checking the sender and his activity on the social media (a human being can’t tweet every minute, but a bot can). The analysis showed that out of the 2,8 million twitter accounts, which had sent the many tweets, 400.000 or about 15 percent were likely to be crafted by bots.

In 2014, Brazilian researchers created 120 so-called twitterbots and set them loose on the internet. Some of them were programmed to search for tweets with a certain content – which could have been climate sceptical messages – and retweet them, while other bots also made comments themselves. The bots are so good at dealing with social media that other users often find it hard to distinguish them from ordinary people. Over a month, twitter themselves revealed only 31 percent of the Brazilian bots, and 25 of them got at least 100 followers over the short period. Only 50% of all twitter users manage to get so many followers in the same timespan. In addition to captivating headlines and slogans, the bot-generated tweets also contain links to websites with more detailed information, and used in conjunction with targeted advertising, they can be an effective tool for spreading fake news. For example, climate sceptics can over and over again publish small shock waves that challenges the credibility of climate scientists.

Fact checking shall restore distrust to the media

In a US survey from 2017, Professor of Communication Kelly Garrett of Ohio State University concludes that about half of the population primarily trust their own gut feeling and do not feel they need factual information to form their own opinion. This group also denounces the established media, claiming they manipulate the news according to their political belief, and it is first and foremost these people whom the false news target. According to a survey of 70,000 people in 36 countries, published by Reuters in 2017, the distrust of the media is greatest in countries such as the United States, Italy and Hungary, where the political debate is highly polarized. Here, less than 40 per cent trust the media, while trust is greatest in Finland by 62 per cent.

The survey also shows that people generally rely more on the media they use and get their news from, while they question the credibility of other media. In most countries, the difference is relatively small, whereas in the United States it is significant. Here, 53 per cent trust their own favorite media, while only 38 per cent trust the media as a whole. It makes the American people an easier victim of fake news because they are not sceptical about media such as Breibart if they normally read it. In an effort to make the climate debate more sober and increase public awareness that global climate change is real and requires us to do something to counteract it, climate scientists are trying to uncover all fake news. Besides Snopes.com, one of the initiatives is the organisation Climate Feedback.

They have a small army of climate researchers with at least a PhD, who searches the internet for articles about climate change and reviews them. Any factual errors and comments are highlighted in the original article on the organisation’s website along with links to credible background information and sources. People who have read a climate article online and are in doubt as to whether the arguments are solid, can then access the website and get answers to their questions. In addition, Climate Feedback also informs the author and editor of the article if they find any problems with the credibility of the content. Facebook and other social media have also gradually started to acknowledge that they have some responsibility for the many fake news that thrives on the internet.

Among other initiatives, they will use advanced algorithms and artificial intelligence to spot and delete fake news, but part of the problem can also be solved by very simple methods. For example, facebook revealed the Russian influence on the American presidential election campaign simply by noting that the many posts came from accounts that were set to Russian as local language, and typically were posted when it was day in Russia and night in the United States.

But until the social media manage to take control of the bots and tiden up the well-organized spread of questionable claims, you must equip yourself with a good deal of critical sense and learn to spot the fake news.

Fake news flourish abundantly in science. Interest groups may distort research results in the media so they apparently support their attitude, and scientists can exaggerate the potential applications of their research to stir public interest that might lead to more founding money. Knowing the pitfalls where science news become problematic is essential when you get an idea for a story or evaluate your sources. The EUSJA journalistic team has made this step-by-step guide to help you assess whether a news story is scientifically substantiated or not and whether the results are worth believing.

By Gorm Palmgren, freelance science journalist writing for EUSJA’s NUCLEUS project

1) DOES IT SOUND TOO GOOD TO BE TRUE?
A headline should always be exciting and make the reader want to dive into the text, but if it sounds unrealistic, it should raise a red flag. Perhaps the author of the article does not understand what he writes about, or she may be trying to manipulate you to have a particular attitude. President Trumps infamous tweet “The concept of global warming was created by and for the Chinese in order to make U.S. manufacturing non-competitive” is a very bold statement, which really has to be backed by solid arguments. The tweet goes unaccompanied, so there is good reason to be sceptical.

2) DOES THE SENDER HAVE A PERSONAL INTEREST?
Although media, interest groups and businesses can be dead serious, they often have a strong interest in what they are writing about. It can affect their arguments and conclusions or it might influence what they choose to talk about. If the oil and coal industry is behind a story that calls off global warming, there is good reason to be sceptical. But if the same news comes from Greenpeace, the supporting evidence is probably paramount since it made environmentalist group change their opinion.

3) IS THE ARTICLE BASED ON OVERCONFIDENT POSTULATES?
Research results are never completely clear, but can be explained in many ways. Therefore, you should always be critical of results and conclusions that seem arrogant in their postulates. If the article itself is very skewed and does not question its own claims, then the content is all the more difficult to rely on. A claim like “The results show, that the new drug has no side effects” can by definition not be true and should make you question the validity of the whole article.

4) ARE THE ARGUMENTS BASED ON SOLID GROUND?
When researchers are to analyze their own test results, they often use statistics, and it usually only makes sense if there is a large dataset behind. If a new herbal drug against the flu is tested on five persons and all get cured within a week, the researchers – or the journalist reporting about the study – might be tempted to claim “100 percent efficiency”. But it would be more fair if they kept silent and repeated the experiment with much more patients as well as healthy controls.

5) ARE THE STATEMENTS CHALLENGED?
An article, based solely on a single research result or source, is completely unchallenged and can easily resemble a promotion of a particular point of view. It is always good if the claims are held up against other researchers’ comments which might put them in a critical light, or if the results are compared to other research that may reach different conclusions. A well balanced source allow you to better make your own opinion or just realize that the truth is complex.

6) DOES THE ARTICLE DESCRIBE THE SCIENTIFIC METHODS?
If the article describes how researchers have achieved their results, it is much easier to judge whether the results are trustworthy or just hot air. If the researchers’ methods are not mentioned, it is very difficult to evaluate the facts and you can easily be fooled into a false or biased conclusion. If the claims are based on a poll, it is important to look critically at the wording of the questions, who has been asked and how they have been chosen.