Data points and where it doesn't point

Twitter was something I did not deign to use for a very long time. I probably poked and prodded it around the time it came out, and then later and could not really "get" it. Why express yourself in such a limited number of characters? Since I'm writing this in a now rarely used format, the blog, you can imagine that the limitation was not appealing. At one stage in my life, I used Facebook updates for what would be considered tweets; and then since coming back from China, I haven't really had the desire to share short messages broadly. 

I still don't but oddly Twitter has come into my life. It happened in 2019 when the SkyCity Convention Centre caught fire and I realised that the fastest way to follow the updates from civil defence was to "follow" civil defence. Late 2019 it went on and 2020 brought a whole lot of new reasons to keep abreast of current happenings. But that is not Twitter's main flow of information - majority is just people expressing themselves about anything and responding to others and because of its fast-moving nature, and the smaller size of journalistic staffs at media companies, it has since become a proxy vox pop for what "people" think. Unfortunately, Twitter represents and amplifies only a subset of humanity's views and so there is a lot of power in the influence of some. It is rather depressing at times to read because you do start to question yourself how common some opinions may be when you see it expressed and a chorus sing in support of it. As well as the pile-ons of criticism that some people get.

But it is also an incidental observational study of humanity, too. It really does just illustrate how confirmation bias works, where humans naturally look for things to support their viewpoints, and tend to reject things, often vehementally, that challenge them. Probably the one phenomenon that you can see even more easily on Twitter than in our everyday interactions is the state of being "triggered". The linguist in me likes the fact that this term, in the sense that we use it today is very new. It really got a push in the tumultuous year of 2016 - think Brexit and Trump. Google Trends show it nicely:

My understanding is that "triggered" was an extension from the use of "trigger warnings", a term which only started to appear in the 2010s with a big increase in usage in 2014 and 2016 and 2019. Though there is almost certainly a cultural, zeitgeist feeling to these terms, that could consign them to the lexical dustbin, I do think they have a certain utility and express something quite tangible and unique to other terms. 

For some definitions, a "trigger warning" is a label, sign or verbal note before something begins to warn the participants or audience that there may be a potentially upsetting topic exhibited. For example, often before news about suicide, rape or other violence, a presenter may suggest viewer discretion whether to watch or not. This has extended to what lecturers or speakers giving such warnings before discussions. The trigger is a particular theme, topic or word, that could cause psychological distress to some people. They would be "triggered" as a result. 

Triggered in the sense above is quite understandable - someone with suicide in the family, or who has had a history of suicide should be given some time for deciding whether to participate or how to participate in a discussion. However, trigger warnings got associated with liberal college campuses in America, and the more culturally conservative side saw them as a weakness, that academics who should be preparing for the "real world" actively avoiding it.  

From this it got the extended meaning of just reacting strongly to a word, idea or topic. In the term of the previous US president a term, Trump Derangement Syndrome (TDS), humorously emerged to describe how the liberal side of the American spectrum (and most of the non-US population of the world) became enraged or agitated by almost any utterance or action by him. Triggered, if you will.

Why do I raise this with respect to Twitter? People get quite, well, triggered with any suggestion or mention of one of:

- lockdown

- vaccination

- masks

- "freedom"

- mandates

- Government control

and often cannot move past them in order to logically weigh up further nuances in what is being raised. 

It calls to mind the hair-in-my-soup kind of reaction where basically anything else served is essentially replied to with: "But there is a HAIR in my SOUP!" Possibly due to some cognitive shortcut they have associated two things, for example, vaccinations with pollution, or lockdowns with fascism, that once something is mentioned; or some form of ideological hygiene where one cannot brook a scintilla of an iota of connection, like the religious conservatives reaction to contraception or abortion. While many people have now adapted to lockdowns, others take the idea, let alone the life under it, as an affront to their dignity, common sense and any sense of goodness in the world. Worse, they could feel it as a threat, that this is the start of the slippery slope toward some form of a New World Order, authoritarian takeover.

When there are strong reactions like this, it would seem any example becomes confirmation, and any counter-example a red rag to charge. In this particular phase of the pandemic, New Zealand and Australia have swung back into global focus for our late winter Covid troubles. While hailed at various times for our apparoach, the Twittersphere from all over the world has weighed in on our outbreak, especially the schadenfreude of the anti-lockdown, "Plan B" factions. For many of them, from even New Zealand's first outbreak and lockdown, they voiced cynicism about eliminating Covid, and that the cure of lockdown was worse than the disease. The blushes and crowing they received for the long stretches of Covid-free here probably burned for quite sometime while everyone was having a lovely, free summer. But even then their data-point de jour was always: "What about Sweden?", Sweden whose two waves of outbreak left it with a considerable body-count although not the epic numbers seen in Western, Eastern or Southern Europe. (But don't ask them about the rest of Scandinavia because the Finland and Norway kick its arse...) 

The data point is thus the shield to protect the user from further thought, because it supports that gut outrage - but the problem is that data points are just points devoid of their place in the trend, or even their place within the trend, are they close to the the correlation line, or an outlier. The art of abusing a data point is the taking of an outlier, or just a decontextualised form of evidence and neutralising any character from it apart from the point that you wish to make. For the Sweden data-point, the point is that with lax measures they achieved a better than average outcome for most countries. But the idea of laxness is often seen as an absolute sense: Sweden, lax; NZ, strict. There was a measure of stringency for Covid-19 measures established, and a chart of the two countries are below from Our World in Data:

Clearly, on average we have had it lax with brief patches of strictness and clearly our peaks were higher than anything Swedes had to endure. Of course, New Zealand itself is a data point raised by those in favour of short, sharp responses which again is not appropriate because New Zealand has some very nature advantages such as being an island, one government over the whole to ensure consistency of messaging and approach. This isn't possible to too many other countries, but nor is it a surefire success.

Sweden, to be clear, has some clear characteristics. They are culturally more distant, have Vitamin D fortified staple foods and were blacklisted from many countries because of their laxness, which in turn meant that they were not seeded much with other variants of concern. It might also be quite possible that being ethnically homogenous might serve some "data points" better than others for some variants. A kind of Russian roulette to see who is affected or "dodges the bullet". In a more diverse country, there might always be a portion of the population vulnerable.

During the verbal meleeing for this particular Delta outbreak in New Zealand, there have been some further examples. Two Australian states have given up pursuing elimination (Victoria and New South Wales), with the Prime Minister saying that Delta has changed the rules and it is impossible to pursue zero Covid and now trying to cow two zero Covid states to loosen up! There have been some countries such as Vietnam and Fiji that after success in elimination in 2020 have been overwhelmed by Delta in 2021. Name any of these and you would think New Zealand too should give up. But China and Taiwan have both suppressed outbreaks of Delta. Really raising data points here have different sides raising their own trump cards and thinking they've won. The argument may be resolved for New Zealand in a week or so if Auckland can have start having "doughnut days" (days with no cases). 

There has been something of a strawman argument that has accompanied the anti-lockdown choir that zero-Covid was something any place was pursuing forever. Even Scott Morrison's comments were in that vein: "Any state and territory that thinks that somehow they can protect themselves from COVID with the Delta strain forever, that's just absurd," which even Australia's plan has never been, nor has it been New Zealand's. Yet, zero-Covid is attacked thusly.

The latest "what about" data points have been Israel and India, but for opposite reasons. Israel was a poster child for vaccination early, with a swift, corporately owned roll-out which had a large proportion of the population vaccinated before most of the world got out of bed. It has recently had its own Delta outbreak and people are dying. Also research there has shown that vaccines have had a waning effect over time. Yet, while their name lingers as the "leader" other countries have long since passed them:

The problem with taking Israel as a data point is that there is plenty of idiosyncrasy in their situation: early adopters (and thus had their Pfizer three weeks apart), exposed to the Delta variant at a time when the earliest vaccinated, the elderly, have the most diminished antibodies, and an ultra-orthodox portion of the population who are not willing to be vaccinated. 

"What about India?" Well, now that we've mentioned Israel, there are many who will compare the rate of Covid for these two countries. India has started to use ivermectin since its Delta outbreak and still has a low vaccination rate; Israel is all on board with Pfizer:

"What about India?" and "What about Israel?" is thus the catch-cry of the Team Anti-Vaxx, Anti-Big Pharma, Pro-Ivermectin. India hasn't had any further Delta burn (relatively speaking). Often the graph above is only showing the segment of time from May 2021 to the present time. This is a bit cynical because the previous trends are quite revealing of the nature of India as a whole: either numbers are under-reported, or Indian nationals are less susceptible to Covid. Smaller populations are more likely to have the huge spikes like Israel above, while big countries like India, Russian and the United States should have lower spikes, anyway, because regions without outbreaks mitigate the impact of those with outbreaks, but all things remaining equal the number should be balanced. Above, even before the vaccinations and Ivermectin, Israel had a massively higher case rate. The data points are thus moot data points.

The final data point is Delta itself. There is a lot of hindsight bias that results from what has been the most difficult variant yet. It's easy to say that an approach was always wrong because Delta came, but that ignores that Delta was not an inevitability. A contagious, yet more innocuous variant was just as possible. Yet conclusions are made to serve the a priori beliefs.