Nupur J. Sharma, in Independent, 15 November 2018 where the title is “The BBC research on ‘fake news’ is shoddy, unethical, dishonest, and actually an example of fake news”
The British Broadcasting Corporation (BBC) has published a report [pdf] that talks about the fake news industry in India. In their report titled “Beyond Fake News”, BBC has reached certain outlandish conclusions like ‘Nationalism is driving the spread of fake news” based on dubious data, incorrect interpretations, bias sample selection and selective information. It was almost as if BBC decided the conclusion of their research and then reverse worked their data to suit their conclusion.
Following are just 5 aspects that prove beyond reasonable doubt that the BBC research is unreliable at best and fake news at worst.
1. Sample size
The BBC “research” has mentioned its sample size in no uncertain terms. For a publication as reputed as BBC, one would imagine that before publishing any research paper, their sample size for the survey would not only be comprehensive qualitatively but also quantitatively.
The BBC sample size as mentioned in their document is as follows:
As can be seen from BBC’s own document, the sample size used to “interview” based on which they came up with their conclusion was 40. 10 each from 4 age brackets. At first glance itself, even a rookie would say that the sample size is abysmally low for a vast and diverse country like India. If you explore the generally accepted norms for sample selection and sample size, one would realise that BBC is way off the mark and the sample size itself negates their entire survey, research and the conclusions reached thereof.
Survey Monkey, a reputed survey website uses the following thumb rule:
To quote SurveyMonkey:
“So if, for example, 90% of your sample likes grape bubble gum. A 5% margin of error would add 5% on either side of that number, meaning that actually, 85-95% of your sample likes grape bubble gum. 5% is the most commonly used margin of error, but you may want anywhere from 1-10% for a margin of error depending on your survey. Increasing your margin of error above 10% is not recommended”.
According to SurveyMoney, 5% is the most common margin of error. Even if we take a small population size of 100, one would have to interview approximately 80 people to ensure that the margin of error is limited within 5%, according to the SurveyMonkey chart.
To put things in perspective now, India has a population of 1.3 Billion people. There are over 7.83 million active Twitter users in India and are set to rise to over 34 million in 2019. As of October 2018, there were as many as 294 million Facebook users in India. BBC’s survey sample size was 40.
Let me repeat that again. With the entire population of India exceeding a billion and with millions and millions of users using Twitter and Facebook, BBC decided to conclude that ‘Nationalism is the driving force behind fake news’ on the basis of 40 people. BBC’s sample size was 80 people from 3 countries and 40 people from India.
Even if we take the sample size of 1 million people (since the sample size doesn’t grow in the same ratio as population), BBC would have to interview 384 people to stick to a 5% margin or error. I reiterate they have interviewed 40 people.
It suffices to say that the shoddy sample size alone should be grounds to dismiss this report by BBC as bogus, propaganda driven and motivated.
In their defence, BBC has claimed that their analysis is qualitative. There is an age group 19-24, which is a very large and diverse group and is also the main consumer of the internet in India. For such a large group, the sample size is 10, which is half of what’s the minimum requirement is even if one buys BBC’s excuse of justifying their sample size. Though unconvincing, we are including BBC’s response here in fairness, which we don’t expect from BBC.
Even if we accept BBC’s theory, by the law of large numbers and Central Limit Theorem, the bare minimum requirement for sample size is 30 (n=30) to ensure that the sample is equally representative. This is important to ensure consistency and to ensure that all estimators are close to the population value. By the extension of this logic, what was required was that for all age-wise segregation, they should have picked 30 samples individually, for each segregation. Instead of that, BBC decided to pick a total of 40 samples rendering their sample inordinately inaccurate, biased and seriously dubious to be considered for the basis of any serious research on the subject.
This sample was used to determine “why people share fake news”. For such qualitative assessment where a humongous population is being painted with a broad brush, the statistics don’t add up.
It does not matter whether the study is qualitative or quantitative in nature. The early stage of designing a research study involves deciding upon a size of the sample which will render the estimated value to be as close to the population value. Hence, they got the design of their study wrong in the very initial stages of the research itself.
Besides, BBC doesn’t appear to have great knowledge of Qualitative Research. One of the ontological positions of Qualitative Research is that there are multiple realities and that there is no single objective truth. The purpose of Qualitative Research is not to arrive at objective truths that can apply across all situations but to generate hypotheses that could be tested using Quantitative methods. The focus of Qualitative Research is always on the subjects that are selected as participants and great care must be taken to ensure that the participants are genuinely representative of the larger population if one is to derive any meaningful understanding regarding the larger world. If the selection of subjects themselves are not done in a reliable fashion, then the conclusions derived from them won’t be applicable to any other group.
However, unlike BBC, we like to explore a little more and hence, the other glaring flaws in the report must be pointed out as well. However, this counter would only be indicative because if the entire report is to be countered, we would have to write a book that runs into over a hundred pages.
(Note: This point has been updated with relevant information)
2. BBC passed off truth as ‘fake news’ in its ‘research’
The BBC research on fake news is so shoddy, that they have passed on genuine news as “fake news” just to establish the “Nationalism leads to fake news” narrative. In their research document, they have included one such.
This news was covered by OpIndia.com as well. One wonders then if the BBC is passing off real news as fake news and OpIndia is covering the real news, who is the fake news purveyor among the two? But I digress.
The muster of the researchers in this BBC fake news research paper is so abysmal that while the snapshot they have included says “diwaliya” which means bankruptcy, they have transcribed it as “Diwali”, the festival of lights in the following text. Had they known that it means bankruptcy, they’d have recalled that BBC itself had praised the law introduced by Prime Minister Modi as being “good for business”.
3. Doctored data set
BBC has put in some handles as seeders amplifiers etc. of fake news, and selection of these handles itself is based on ‘sources’ that are biased and have often peddled fake news – as we will see later in this article – and this selection is not just biased, but absurd.
Based on “fake news sources” identified by these fake news peddlers, the BBC report makes the following conclusion: “After plotting the sources who have published fake news, on this network map a pattern does emerge, where we see that handles that have published fake news sit in the pro-BJP cluster”
First, BBC has included a Twitter handle called RealHistoryPix as a purveyor of fake news. This handle is a parody handle – as clearly mentioned in their Twitter bio – and often tweets things which are not ‘news’ or makes assertions on current affairs for satirical purposes. For BBC to identify this handle as a purveyor of fake news is simply outlandish.
On page no. 87, they name OpIndia.com, R Jagannathan and Swati Goel Sharma as three of the several accounts that Prime Minister Modi followed who are purveyors of ‘fake news’. The BBC report says, “And of the 30 sources known to have published at least one piece of fake news, that sit in the pro-BJP cluster, the @narendramodi account follows 15 of them.” – this is where OpIndia is listed.
If “at least one piece of fake news” is the criteria, BBC qualifies liberally. Further, the “at least one piece of fake news” is an absurd criterion is what we’d see in the omissions section later in this article, where BBC didn’t care about many outlets that have published various fake news items, not just one.
The BBC then goes on to cover their shenanigans with some disclaimers. They say it is possible that these accounts may have spread fake news based on slippages of journalistic standards or because the ‘entities’ BELIEVE in the truthfulness of the claims. And then the clincher. BBC says “And finally, it is possible, though unlikely, that the independent fact checkers who have identified these sources as having published fake news have made errors”.
In all of this, they have provided no data as to why OpIndia, Mr Jagannathan or Swati Goel Sharma have been branded as fake news purveyors. None at all. The BBC survey conveniently ignores the deliberate fake news spread by their own sources and brands several others fake news purveyors with a disclaimer that they might be wrong.
The only reason one can imagine that these three entities would be included in this list is, that, all three have often exposed the bias and fake news furthered by their “sources” that have selected the handles, like this one by Swati Goel Sharma exposing Indiaspend, one of the BBC’s sources.
I would like to take this opportunity to challenge BBC and ask them to show instances where OpIndia has deliberately spread ‘fake news’. The two times or so that we did make a mistake, we updated our articles and the Editor tendered an unconditional apology. We would like to challenge BBC to show us where the same integrity has been shown by the very sources of fake news that they have considered an authority to help them with their research.
In fact, here is something that shows the lack of muster in the researchers and the BBC report itself.
There is a website called ‘ViralIndia.net’. This website was run by one Abhishek Mishra who was extremely close to Senior Congress leader Digvijay Singh and Aam Admi Party supremo Arvind Kejriwal and had used that website to spread Islamist propaganda and Anti-BJP, Anti-Modi fake news. In the BBC research paper, this website was, on Page 88, was listed in the Pro-BJP cluster. This itself is fake news.
Then, as if BBC was fact-checking its own fake news, on Page 102, this website was listed in the Anti-BJP cluster.
4. Inherent bias in the introduction of the report itself
The introduction to the entire report itself indicates the BBC report’s bias and why they might have concluded that ‘Nationalism is the source of fake news’ in India. In their introduction to the long report, BBC writes: “Today, it is fair to say, the term ‘fake news’ carries a few positive associations. Today it is an inarguably negative term, irrespective of who is using it, though broadly speaking all users of the terms refer to misleading or false information. The term has been used by journalists and researchers in conjunction with words such as ‘crisis’, or even ‘democratic crisis’. As is well known, influential politicians around the world have taken up the term to connote any news that is critical of them and their achievements.”
The text above has been referenced by BBC with a Quartz India article. The Quartz India article, when talking specifically about India, first quotes a CNN-News18 report that talks about Amit Shah joining 18,000 Whatsapp groups ahead of 2019 elections. They connect this quaint, insignificant report to a Times of India article from 2014 that spoke about Social Media being a pivotal campaigning tool for ALL parties. Then, the Quartz report connects the TOI report and the CNN-News18 report to 2 articles on BuzzFeed. Both written by Pranav Dixit, one headlined “Modi’s Political Party Creates Abusive Social Media Campaigns And Breeds Internet Trolls, Claims New Book” (the book was ‘I Am Troll’ by Swati Chaturvedi) and the other, headlined “Viral WhatsApp Hoaxes Are India’s Own Fake News Crisis”.
Pranav Dixit is a writer who, owing to his pathological hate for anything that doesn’t conform to his Leftist views, attempted to dox, stalk and threaten a woman who dared to simply disagree with him and happened to be right-leaning in her political views.
In the first article written by Pranav, which has been referenced by BBC, Pranav talks about a book, ‘I Am Troll’ written by “journalist” Swati Chaturvedi, who, owing to the strong Left ecosystem has won accolades for writing gibberish. Swati Chaturvedi, who abuses incessantly, tweeted tens of times targeting one Rahul Raj’s job. Eventually, she succeeded in getting him fired owing to constant character assassination sans proof. She willy-nilly included his name in her book.
She weirdly cooked up a story about actor Leonardo Di Caprio being invited to an RSS event. She lied. She attributed an image of a girl beaten up by a deranged jilted man to violence in a university called BHU, just to settle political scores. She has been accused of plagiarism. She has lied on national television about her book, which has been quoted by the BBC while promoting her book. She has been slapped with a defamation case for accusing a man of “being arrested for sexual harassment” when he never was. She has jeopardised the safety of intelligence agency officers by irresponsibly revealing their names and identities on Twitter. It is frankly a miracle she wasn’t booked under the Official Secrets Act.
She has spread lies about a simple speech by Amit Shah, the BJP Party President. She has merrily given credence to fake news, from a dubious website, which was later debunked. She has lashed out at the lady speaker of the house simply because she dared to discipline Congress President Rahul Gandhi after his unruly behaviour in the Parliament. Political party AAP spread lies about IAS officers being on strike, Swati Chaturvedi who is being used as a source in a survey about ‘fake news’ repeated that lie and furthered it. She has been accused of featured on a list who were extorting an absconding scamster. She bullied a young NCC cadet simply because the lady questioned Rahul Gandhi’s ignorance.
The BBC report mentions the Cambridge Analytica fiasco as one of the reasons for the fake news fiasco, interestingly, Swati Chaturvedi defended Congress when their name was taken by the whistleblower and even by a journalist who saw the Congress poster on Cambridge Analytica Chief’s wall. She lied blatantly about a BJP-RSS tussle. She lied about the Information and Broadcasting Ministry withholding funds from Prasar Bharati. She maliciously reported a speech by Indian politician Manohar Parrikar. She lied that Kanhaiyya Kumar attacker was a BJP office-bearer. She even spread her malicious propaganda against BJP minister after a train tragedy that claimed many lives at a Congress event. She wrote a lie-ridden, asinine report targeting India’s NSA’s son, Shaurya Doval which was debunked within minutes.
Even the Swedish Embassy cancelled their invite to Swati Chaturvedi after realising her reputation for being a purveyor of fake news.
5. Reliance on sources that are biased
This is where it gets really interesting. On pages 82 to 88, where the BBC research talks about the “fake news ecosystem”, they have set up a neat little malicious conundrum. They first say that they have mapped the clusters that are followed by PM Narendra Modi and Rahul Gandhi. They identified a bunch of “fake news sources” and then drew the conclusion that the Pro-BJP fake news disseminators are far more closely knit than the Anti-BJP fake news disseminators. Notice that they don’t use the term pro-Congress.
How have they identified these “fake news sources”? A large part of their methodology which has been repeated several times through the BBC report is: “As such, our analysis is not about ‘fake news sources’ (as that indicates there are sources who only publish fake news) but about ‘sources that have published fake news’. For the purposes of the Twitter network analysis, therefore, we define two categories, 1) ‘Sources of Identified Fake News’ and 2) ‘Likely Fake News Disseminatiors’. The ‘Sources of IdentifiedFake News’ are established in the two following ways:
A. Identified as fake news by a fact-checking site (one or more of the following: smhoaxslayer, altnews, boomlive.in,factchecker.in) or by the platform(e.g. Facebook/ Twitter)
B. Identified in the qualitative fieldwork by researchers or picked up from the /Unkill thread on Reddit.
Let us look at the sources that BBC has depended upon to identify ‘fake news disseminators’:
The co-founders of AltNews are Pratik Sinha, who is pathologically anti-BJP and anti-Modi and one parody handle called ‘susu swamy’. Firstly, it is interesting to note that BBC would depend on a website that is co-founded by someone who likes to call himself a variant of urine. Either way, to attribute the status of non-partisan “fact checker” status to altnews reeks of an agenda.
When conferring the status of neutral fact checkers to Altnews, BBC never realised that when anti-BJP voices spread fake news, Altnews steps in to justify the same. Suddenly, it becomes “easy to fall prey to fake images” as “one is outraged about an incident”. Even fake news gets a “context”. It is deemed that these people had “unknowingly” shared a fake image. Worse, these people are made heroes for “admitting” that they lied. And Altnews cannot hide its happiness and is “glad” that the fellow communists have “regretted their mistakes”. We have covered these shenanigans in details here.
While Altnews justifies the fake news peddled by their ideological compatriots, they themselves have often been caught lying blatantly and have seldom corrected themselves. Recently, they sought to ‘fact-check’ a ‘clipped’ video of journalist Barkha Dutt but interestingly, left out the quotes where she was actually giving a context to the genocide of Kashmiri Pandits in the valley. Altnews ‘fact-checked’ a retired major from the Army, but willy-nilly refused to respond to his side of the story. Altnews fact-checked a fake image in circulation without bothering to name the ones using that fake image. The reason was rather simple. The ones sharing that image were their ideological comrades which included Swati Chaturvedi, a source BBC has relied upon. One of their co-founders happily furthers genocidal accounts wishing for death upon anyone who doesn’t subscribe to the leftist cult. The co-founder of Altnews spread lies about ease of doing business rankings. He shared a fake image of BJP leader BS Yeddyurappa right before the Karnataka elections.
The website did some strange mumbo-jumbo analysis to debunk the news of several youths raising anti-India slogans. The Bihar DGP debunked Altnews claims. They are funded by a fake news purveyor, Arundhati Roy (BBC might be reluctant to admit that Roy is a fake news disseminator, but they can read up here). The co-founder doesn’t bother to fact-check even when fake news hits him in the face as long as it conforms to his political bias and the co-founder of Altnews has retweeted fake images in the past.
Factchecker.in is the ‘fact-check’ arm of Indiaspend. Indiaspend itself peddles “research” just as bias as this fake news research by BBC. Perhaps it is no surprise that the two seem to be a match made in heaven. Considering the BBC research is about how fake news gets furthered, it is interesting to note that they have also relied on this website, whose parent organisation’s fake news was carried by some of the biggest names in media, including Barkha Dutt and Washington Post.
Indiaspend has also spread fake news about the rise in sexual harassment cases in Uttar Pradesh. Indiaspend employees too spread fake news about Narendra Modi while writing in other media portals like The Hindustan Times, no less.
Factchecker.in itself has been caught lying in their article about the Power ministry and done a “fact-check” based on misreported and twisted information in the mainstream media. They have in the past ‘fact-checked’ PM Modi’s claims in his independence day speech and ironically, their ‘fact-check’ itself was fake news.
The misinformation, bias, and propaganda by these outlets have been regularly exposed by us at OpIndia, and by alert observers like Swati Goel Sharma, who works with the Swarajya magazine where R Jagannathan is the editorial director.
These are the “sources” based on which the BBC report branded OpIndia, R Jagannathan and Swati Goel Sharma as handles that spread fake news. The BBC’s agenda is clear and it is also evident why these biased “sources” would include the three names as fake news purveyors.
6. What BBC missed
When they analysed the names of fake news purveyors, they have missed several such entities who are known to have political ideologies and spread fake news. Since BBC has taken it upon itself to name several handles as fake news purveyors sans proof based on the say-so of fake news purveyors themselves, here is proof of leftist propaganda websites and how fake news is spread with impunity.
This list is only indicative. If one wants to access all articles that fact-check The Wire and bust their propaganda, please see here.
The publication is known to peddle half-truths and whole lies as a force of habit with a long list to their credit. The Wire, which is being sued for defamation often publish articles that read like Hurriyat press statements. Their article has in the past incited violence. They were recently slammed by the CBEC for peddling fake news. They were blamed by a stand-up comedian for spreading lies in his name. Have peddled ‘data’ from dubious organisations. Was slammed by cricketer Mohammad Kaif was bringing caste into cricket. Has a video series that regularly obfuscates. Its journalists are abusive trolls and people who share the communally sensitive fake news. They have in the past tried to rationalise the 1993 bomb blast. Conducted a hit job against Sri Sri Ravi Shankar. Defended Chidambaram’s 20:80 gold scheme that benefitted private players like scamster Mehul Choksi. The Wire is a portal where the Editors tell their journalists to not out “Dalit activists” who heckled her. Where their contributors are accused of sexual harassment and their journalist’s conduct drove a female journalist to commit suicide. They had shamelessly compared the Indian Army General to General Dyer.
Earlier, Rahul Gandhi had shared The Wire’s comprehensively debunked hit job on Piyush Goyal with some caustic comments without realising his self-goal perhaps. He is no stranger to self-goals too. In fact, not so long ago he had done exactly the same when we at OpIndia has preemptively busted an imminent hitjob against the Finance Minister, Arun Jaitley. After our article, ‘The Wire’ had conceded that “there was not much of a story to pursue after all“, which is why they had dropped the story after receiving a response from Jaitley and his daughter’s firm. Mr Rahul Gandhi had then proceeded to plug the very article where ‘The Wire’ had admitted to no impropriety on the part of Arun Jaitley, in a bid to tarnish his image. We had then also speculated whether Rahul Gandhi actually knew of the imminent hitjob. They had conducted a hit job against Amit Shah’s son Jay Shah hitjob which was debunked by us comprehensively and the portal responsible for that hitjob is facing a 100 crore defamation suit, in which the court has already mentioned that a prima facie case exists against the portal. They had also recently conducted a hit job against Amit Shah which was debunked by us.
It is apparent that The Wire’s editorial standards are near absent. After all, it is run by an Editor who had earlier shelved the Robert Vadra land scam story for 7 months for no good reason. This latest lie is just a reflection of their integrity.
It is indeed interesting that the BBC would talk about ‘fake news’ without talking about the mainstream media has institutionalised the conjuring up and dissemination of fake news. OpIndia.com itself has fact-checked hundreds of such instances where the mainstream media has either lied blatantly, twisted facts or peddled a fallacious narrative owing to their political affiliation and/or ideological agenda. Those articles can be accessed here: Media Lies List.
In the BBC research paper, they observe that audiences following BBC News are 40x more likely to also consume NDTV.
That doesn’t really surprise me because BBC has not only missed taking cognizance of the heaps of fake news spread by NDTV but also ignored the lies spread by BBC itself (this “fake news research” apart).
Just recently the BBC had retracted an article they had published about a Sikh police officer getting death threats because he saved a Muslim man. These claims were summarily rejected by the police officials.
BBC’s inherent bias against the BJP is really no secret and hence, the conclusions of this ‘fake news research’ is no surprise either. BBC had shared an old report about EVM hacking just a day after BJP had swept the Uttar Pradesh elections and had just a few days ago ‘expressed regret’ over their Kaziranga documentary, also exposing their anti-India bias. This apology came after BBC was banned from filming in all national parks and sanctuaries in India, in 2017, for 5 years.
Congress and allies
The BBC loves “liberals” and any political ideology that conforms to their Left bias. They haven’t even bothered to include the institutionalised fake news that is peddled by Congress functionaries and even Congress President Rahul Gandhi. In fact, the entire fiasco is rather amusing because Congress supporters had used a fake BBC article itself to sing songs of praises for Rahul Gandhi. BBC had been a mute spectator then as it is now.
BBC in its report says: “We have also used the term Twitter ‘amplifiers’. These are accounts that disseminate a high volume of content (defined as greater than 10,000 tweets), at least some of which is fake news, to a wide audience (defined as greater than 10,000 accounts).
I would like to point out to BBC that Rahul Gandhi himself has 7.8 Million followers and that BBC has reached the conclusion that “Nationalism drives fake news”, without taking into account the Left lies is an abomination.
An indicative list of Rahul Gandhi’s lies:
- He blamed BJP/RSS for Periyar statue being vandalised. Fake news.
- The French government and the Dassault CEO themselves have called Rahul Gandhi’s fake news on Rafale.
- He lied about seeking permission for Mansarovar Yatra. He claimed the permission was denied by MEA. Fake news.
- He lied about Tata’s Sanand factory in Gujarat.
- He lies about the action taken against Shell companies post demonetisation.
- He shared a story by The Wire which admitted that there was no wrongdoing on the part of Finance Minister Arun Jaitley. He used that article to allege impropriety. He was debunked multiple times. He continued to lie.
- Congress party lied profusely after they deleted a Facebook ad of ‘Modi hatao desh bachao’ in Pakistan.
- He lied about Congress signing a deal with Dassault when in power.
- He lied, repeatedly, about the Statue of Unity being ‘Made in China’.
- There is a criminal defamation case filed against him for lying about Ahmedabad co-op bank.
The list is frankly endless. These are just 10. There are several which can be accessed on OpIndia.com.
The ridiculous sample size should be the first indicator that the BBC research is bogus. Everything else, from their selection bias, to terming others ‘fake news purveyors’ based on the inputs of fake news purveyors themselves, points to the fact that BBC had already made up its mind as to what the conclusion of the report must read.
BBC had invited me to be a part of their panel discussion on fake news. Interestingly, the findings or the existence of this report was not informed to me at the time. I had refused to be a part of a panel that had fake news purveyors discussing fake news and also because I was deeply aware of BBC’s inherent bias. I was told that they want an OpIndia.com representative because they wanted to be ideologically fair.
The panel took place on the 12th of November. Clearly, this “research” had been in the pipelines for months and they had already decided, based on extremely shoddy data and methodology, that fake news purveyors were actually the harbingers of journalistic integrity and that the fake news emanates from primarily one side. The fact they asked for my presence was clearly, a mere formality so they can continue to wear their cloak of neutrality while peddling the worse form of fake news and propaganda.
BBC has proved itself to be a biased, leftist rag. I won’t mince my words because they have clearly not minced theirs in the shoddy propaganda literature that they call “research”. I will have no part of it now or in the future.
One can very well argue that this ‘study’ by BBC itself is a perfect example of how fake news is created and spread through a network of co-conspirators. A study that fails statistical muster has obvious errors, is biased in its approach, helps in the propaganda of a particular political ideology is being cited by every other media outlet as some scholarly work by impartial and ethical researchers. The irony can be missed only by those who have decided to close their eyes to the truth.
Next time BBC indulges in some real research, they should include their own shoddy journalism as one of the examples of fake news.
Editor, OpIndia.com since October 2017
May Bulman: “UN condemns UK government’s ‘mean-spirited and callous approach’ to poorest, in damning report,” November 2018, https://www.independent.co.uk/news/uk/home-news/un-report-uk-poverty-austerity-government-cuts-special-rapporteur-benefits-callous-a8636901.html