Africa Growth Pilot/Online self-paced course/Module 3/Avoiding undue weight - part 2

To sum up, what should we do? What is *due* weight? Articles on the encyclopedia should reflect *the majority views in reliable sources*. Reliable sources! So it's not enough that a lot of people believe something, when there are facts.

To take another example, a medical fact, look at this sentence: "Although millions of people believe one can contract HIV from kissing, medical science sources all agree HIV is not transmitted by kissing."

This is an example of *due* weight, of a good encyclopedic sentence: It acknowledges the fact, and it *is* a fact, that millions of people believe you can get HIV from kissing. That is a fact. But the fact millions of people believe that, does not make it true! It's *not* a medical fact. And the people who have the relevant expertise, the medical professionals, all agree the virus just does not transfer that way, no matter how many stories you hear about "my cousin's neighbor got it from kissing". And maybe some of you have heard that sort of story: people who swear that they got it from kissing or whatever. It doesn't matter. The *relevant* sources here, the relevant view on such a question, is the view of *scientists*, of people who studied this, because it is a medical question. If it were about football teams, maybe football people would have the relevant opinions, as we've seen earlier.

Or maybe there is just no way to establish what is or isn't the case. Like the question of who is the greatest of all time. There isn't a factual answer, but on the question of how HIV is transmitted, there is a decided medical finding. And this would require sources, of course. A sentence like this, when we assert that "medical sources all agree HIV is not transmitted by kissing", would actually have to have quite extensive sourcing, not just one scientific paper. We would need something like a *meta-analysis*, which is a kind of scientific paper that compares and collates the results of multiple studies, and a meta-analysis could say something like "across 130 different studies, none of them has shown contraction by kissing, etc.", then we can support this sort of claim.

And the principle is that extraordinary claims require extraordinary proof. The more momentous or sweeping your claim is, the higher the standard of proof needs to be. I haven't bothered with specifying the citations here, because my point here today is the tone, and we will talk about sourcing in module four.

The coverage we give should also be *proportional*. But before that, there's another question in the chat: one may ask *who* believes that HIV can be contracted by kissing? Yes, absolutely. So when I say "millions of people believe one can contract HIV from kissing", that *also* needs to be supported by a citation. And if I don't have some study that shows that, say, in a survey of people 30% of them said that they believe you can get it from kissing, and that 30% in the population that was surveyed is about 5 million people -- if I don't have such a survey, I *shouldn't* say that millions of people believe. That's not something I can just say. Absolutely. We need a citation for that as well. Thank you for the question.

So, when we cover things, we should also be proportional, meaning we shouldn't *overemphasize* some minor aspect at the expense of properly covering major aspects. Some of you may know Michael Phelps, the American swimmer; look at this opening paragraph for a Wikipedia article about Phelps. If this *were* the opening paragraph (it isn't).

"Michael Phelps is an American former competitive swimmer and the most decorated Olympian of all time, with a total of 28 medals. He has encountered legal trouble on several occasions. At age 19, he was arrested for drunk driving. On February 2009 he had a photograph of him using a bong. And then in 2014 he was arrested again, this time on charges of drunk driving and speeding."

If this were the *opening* paragraph of the article about Michael Phelps, that would have been *undue* weight, even though it is 100% factual. I hope you see what undue weight feels like. This is one of the most remarkable human beings ever in his field of swimming! However much of an irresponsible citizen he is with drunk driving, etc., he's an amazing swimmer and *that* is why we have a Wikipedia article about him.

The Wikipedia article should focus on what he is exceptional at, not the fact he was arrested for drunk driving. It should mention it. It is a fact. But it shouldn't be the majority of the opening paragraph, right after only two lines about his exceptional swimming career.

In fact, the article should start by saying he is a competitive swimmer, the most decorated Olympian of all time, and he's done this and won that, et cetera, et cetera. Then there should be sections about his early sports career, and then his Olympics performances, and then after the Olympics, when he was coaching or whatever, all of that. And *then*, towards the end, there may also be a section called "Personal Life" or something like that, where, yes, you would mention that he was arrested, that he was smoking a bong and it went viral. This is verifiable information that can be in the article. But if it were an *opening* paragraph that focused mostly on his trouble with the law and said *almost nothing* about his remarkable athletic career, that would be *undue weight*.