Another dot in the blogosphere?

Posts Tagged ‘clickbait

A local news report seeks to clickbait with its headline instead of educate. Again.

The number 129 might seem like a large number. It actually represents just 0.03% of all students in Singapore (primary to pre-university levels). That is part of the context.

The other part is that students are actually safer in school due to safe management measures [1] [2]. The distancing, hygiene, and modified classroom practices [3] mean that students are probably at greater risk in transit to/from school when they interact with adults who are less fastidious.

Such context provides quantitative and qualitative arguments for educating the general public. Such arguments make us smarter and more confident of our actions. Reading clickbait keeps us in the state of ignorance and fear.

One aspect of modern information literacy is not just reading beyond the headline, it is also about reading wide to other valid and reliable sources of information. Then another element of information literacy should kick in: Evaluating the original article for its worth.

How do you balance the need to create a headline good enough to get readers to click through and getting an important message across? These should not be on opposite sides, but they are in a CNA news article.

This was the tweeted headline from CNA (screenshot below, in case the original tweet is deleted).

The actual article reported this:

First, ask yourself: How many people bother to click through, i.e., read beyond the headline?

Next, if readers do not read the article, they are left with the information that there are at least 2,700 reports of adverse vaccination effects among 2.2 million doses.

The potential impact of the headline is the attention paid to the 2,700 reaction cases. This creates or reinforces fear that fuels vaccination hesitancy. 

How many then learn that the adverse effects were classified into not-so adverse (common reactions) and actually adverse (serious reactions)?  The latter was represented by 95 cases.

That number of cases is 0.004% of doses administered (95/2,213,888 x 100). The article stated 0.04% which is 10 times higher. The same article has a table which reports the correct figure of 0.004%. The percentage in the main body of the text and the table do not match.

Finally, how many rationalise that 0.004% is a very small incident rate? How low is this chance? You have a 1 in 25,000 random chance getting a severe reaction to vaccination.

How unlikely is 1 in 25,000? I found a summary site of statistics maintained by someone who mined NSC and CDC data. If we were in the USA in 2002, each person had a 1 in 25,000 chance of being murdered with a gun.

If that is hard to relate to, then you get my point. The tiny chance and the large number of doses are difficult to rationalise. Suffice to say that the chance of reacting severely to the vaccination (or being gunned down) is very small.

Think of it this way: If you were in the USA and not terribly afraid that you were going to get shot, you should not be afraid that you are going to react severely to the vaccination.

The issue that writers and editors of newspaper headlines do not seem to understand is human psychology. People tend to focus on the part of the headline that screamed “reports of suspected adverse effects”. The headline also includes the initialism HSA, Singapore’s Health Sciences Authority. So it might come across as a warning. The number of cases could have been 27 or 270, but the focus would have been on the authority and the adversity.

The messaging is important. Recipients have a right to know the possible side effects of the vaccination. The HSA was transparent with its statistics. However, the news agency was irresponsible with the clickbait headline and the wrong calculated figure of the severe cases in the main body of its text.

Long story made short: The Media Literacy Council (MLC) of Singapore was responsible for propagating wrong information. It declared that satire was a form of fake news.

Satire is not fake news. This news article cited two prominent individuals who have said so.

Law and Home Affairs Minister K Shanmugam said earlier in May that the Protection from Online Falsehoods and Manipulation Act (Pofma) targets only false statements that distort facts and not opinions, criticisms, satire or parody…

Associate Professor Leong Ching, dean-designate of students at the National University of Singapore said in a public post on Facebook that “satire is NOT fake news. It is exempt from POFMA”.

The MLC apologised (weakly) and removed its Facebook post and misleading graphic saying that it “gave the wrong impression that satire was fake news”.

Impression? The graphic made a clear statement — Header: Types of Fake News; sixth example: Satire.

Just apologise humbly and sincerely instead of using words that try to save face. If not you give the impression that you are neither sorry nor better informed.

Speaking of being better informed, clickbait is not necessarily fake news. If it was, most BuzzFeed headlines and some Today paper tweets are fake news. I am referring to sensationalist top 10 lists and gossip about celebrities and their kids. These do not count as fake news in my books. Drivel is not news.

Rant over. Viewed through an educator’s lens, this incident reminds me that an authority is not the same as an expert. Both can get things wrong, but an authority sees that as weakness so it is reluctant to admit it. This erodes trust.

There is another lesson. An agency might mourn the madness of a mob. But change the circumstances and we get the wisdom in the crowd. A small group of people in an authority can suffer from groupthink more than a large, loosely-connected group of people.

If you read my reflection to the end, the clickbait title worked. Was there fake news or bad information here?

Video source

The video above has a clickbait title — this one weird trick will help you spot clickbait.

The examples highlight not one but three strategies when evaluating clickbait titles of news or video reports:

  1. Drawing a line between cause and effect
  2. Understanding the impact of sample size on reported results
  3. Distinguishing between statistical or scientific significance and practical bearing

Using Betteridge’s law of headlines, The Guardian published an article titled: Could online tutors and artificial intelligence be the future of teaching?

The short answer to any such headline is no.

The longer answer is that modern online efforts provide educators with lessons on how to teach so that learning happens more optimally and meaningfully.

For example, data from a company called Third Space Learning and University College London revealed this:

An early analysis found, perhaps unsurprisingly, that when tutors speak too quickly, the pupil is more likely to lose interest. Leaving sufficient time for the child to respond or pose their own questions was also found to be a factor in the lesson’s success…

The lesson about teaching that focuses on learning: Give learners opportunities to interrupt and intervene.

The longer answer also focuses on whether such efforts will make humans irrelevant:

Hooper agreed that the aim is not to replace teachers with robots. “There’s a slightly dubious conversation about how AI will make humans irrelevant, but it’s not at all about replacing humans,” he said. “Our whole belief is that for children disengaged with the subject, who are lacking in confidence, people is what matter. An algorithm can’t provide that.”

Even well-meaning teachers sometimes get in the way of learning. Whether you like or realise it or not, it is about focusing on the learner and learning, not the teacher and teaching. The latter are means to the former.

Ambar said that maths used to make her anxious, but since starting the weekly tutorials in Year 5, she has started enjoying it. “When they give you horrible sums, they help you,” she said. “I was scared to do it, but it was actually fun.”

If we focus on the who and how of learning, we will hear more stories that end like this.


Usage policy

%d bloggers like this: