Small samples, amazing results

Подписаться 114 тыс.

Просмотров 43 тыс.

50% 1

Jacklyn Dallas ('Nothing But Tech'), Beryl Shereshewsky and Alec Watson ('Technology Connections') face a question about a highly efficient population study.
LATERAL is a weekly podcast about interesting questions and even more interesting answers, hosted by Tom Scott. For business enquiries, contestant appearances or question submissions, visit www.lateralcast.com
GUESTS:
Jacklyn Dallas: ‪@NBTJacklyn‬, / nbtjacklyn
Beryl Shereshewsky: ‪@BerylShereshewsky‬
Alec Watson: ‪@TechnologyConnections‬
HOST: Tom Scott.
QUESTION PRODUCER: David Bodycombe.
RECORDED AT: The Podcast Studios, Dublin.
EDITED BY: Julie Hassett.
GRAPHICS: Chris Hanel at Support Class. Assistant: Dillon Pentz.
MUSIC: Karl-Ola Kjellholm ('Private Detective'/'Agrumes', courtesy of epidemicsound.com).
FORMAT: Pad 26 Limited/Labyrinth Games Ltd.
EXECUTIVE PRODUCERS: David Bodycombe and Tom Scott.
© Pad 26 Limited (www.pad26.com) / Labyrinth Games Ltd. 2023.

Развлечения

Опубликовано:

12 сен 2023

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 82

@NBTJacklyn 10 месяцев назад

Thank you so much for having me Had so much fun!! Such an honor 🙏🏻

@richardl6751 9 месяцев назад

You look like Jana from We Got The Chocolates.

@iamjamieq 10 месяцев назад

So awesome to finally see Alec join this show! Technology Connections is by far one of my most favorite RU-vid channels!!

@17Blower 10 месяцев назад

Dishwashers will never be the same for me

@iamjamieq 10 месяцев назад

@@17Blower Or heat pumps! My previous house had a heat pump, but my current one has a gas furnace. We put solar panels on the roof and I'm waiting to see how they do for a year and then likely putting in a heat pump instead. Very much inspired - along with my previous experience - by Technology Connections.

@drawswithcrayons481 10 месяцев назад

Came here to say the same!!

@Systox25 10 месяцев назад

Both went together on a trip to hell… literally!

@VoIcanoman 10 месяцев назад

I knew this immediately. And it's not just tobacco and booze...you can measure opioid and other narcotic metabolites - this is one way that we can know just how many more painkillers - both legally- AND illegally-obtained - people in America and Canada use versus other nations (tracking prescriptions only tells you about the legal use).

@marcusellby 10 месяцев назад

This is done in most countries to measure trends in drug usage

@margaretmcnarry4298 10 месяцев назад

Wastewater was my very first thought - maybe because I had just been checking our local Covid levels!

@pieterfaes6263 10 месяцев назад

Checking wastewater is very handy to estimate consumption of various things, like antidepressants, contraceptives, drugs... You screen for the quantity of metabolites (what the human body breaks a compound down to before excretion) of a specific compound in the consumable. And as said in the video, estimating the spread of (some) viruses can be done as well.

@PsyKosh 10 месяцев назад

Woo, I got this one pretty much immediately. Noticed 45 was not 45 people... and fairly quickly mentally went to wastewater samples. (Reason: I'm still covid-cautious and have been obsessively checking biobot for quite a while now, so the idea of learning overall population trends from wastewater data was something that was already on my mind)

@RFC3514 10 месяцев назад

I did think of sewage (it's used to estimate the consumption of some drugs), but I'm surprised alcohol readings from waste water are remotely reliable, with all the different microorganisms consuming (and producing) it.

@programista_html8888 10 месяцев назад

It may be that not alcohol itself is being checked, but the levels of alcohol metabolites, which won't create on their own, only in a human body.

@bloodalchemy 10 месяцев назад

My first thought was 45 hospitals, lung cancer and liver failure. There are many studies on the correlated chance of lung cancer per pack of cigarettes a year, same with alcohol and liver failure.

@apurvabhure3376 10 месяцев назад

This is fascinating, I'd no idea that you could detect things like that so far down the line from the actual person!

@57thorns 10 месяцев назад

They can also see what narcotics is used in what parts of a city (areas as defined by waste water collection).

@loddude5706 10 месяцев назад

Homeopapee? : )

@bloodalchemy 10 месяцев назад

Yep, there are a number of medical ethics arguments going on right now trying to figure out how far down from a source household testing can be allowed to stop individual identification and breaking medical privacy compared to being close enough to get useful test samples for any number of contagious diseases.

@MrDannyDetail 10 месяцев назад

Alcohol makes sense to me (as long as they accounted for possible outflow of waste alcohol from breweries, bottling plants and labs etc), but nicotine ending up in either pee or poo, or maybe both, somehow surprises me.

@sirBrouwer 10 месяцев назад

@@MrDannyDetailbesides the small amount your body Keeps in it self. Everything else that comes in to it will leave.

@bzqp2 10 месяцев назад

My 1st thought was a metaanalysis sampling from 45 other papers, each having millions of people analyzed.

@ojaott 3 месяца назад

Tom Scott's wide smile at Beryl's casual mention of pee at 3:04 😂

@igorbednarski8048 10 месяцев назад

an average smoker smokes like 10-15 a day, a nonsmoker smokes 0, something like 30% of Chinese are smokers, so an average of 3.1 per person per day sounds about right

@Jupiterninja95 10 месяцев назад

Here's my guess at the start: The sample size was 45 convenience/liquor store owners in China, or something like that where they get a report of the overall inventory for that region EDIT: I guess I got the Alec treatment haha

@nick140pk6 10 месяцев назад

Love this show so much listen to it all day at work.

@dovos8572 10 месяцев назад

btw that works with the different types of drug consumtion too. you can measure how much meth and coke and so on is in use inside a city by checking in the waste water.

@zigarettenbruch6999 10 месяцев назад

Damn... I was thinking the same thing as Alec.

@drapen1836 10 месяцев назад

I was so happy to see Beryl.

@TakanashiYuuji 10 месяцев назад

I love the guests on this show!

@arjovenzia 6 месяцев назад

every few years we get a public report that lists drug use by region, so you get new articles saying which region are the biggest meth heads, stoners, soberist etc. with headlines such as "meth captial of australia"

@germimonte 10 месяцев назад

I would have guessed meta-study, the 45 samples were other studies or census

@Trygon 2 месяца назад

This is also how we learned Seattle is the most baked place in the world by an order of magnitude.

@Alsadius 10 месяцев назад

And here I thought it'd just be sales data from 45 of the biggest alcohol and tobacco companies.

@MountainHawkPYL 10 месяцев назад

I got this one right away.

@_lukedunlop 10 месяцев назад

Webcam definitely on

@MisterM2402 10 месяцев назад

Reminds me of the song from Scrubs "Everything Comes Down to Poo". Except, flushing it down doesn't flush away the answer because then you just go check the water treatment plants instead.

@jucom756 10 месяцев назад

I'm at 2 minutes in the video and i'm guessing they went to the companies that sell the stuff and just divided by the populace they are in the region of influence of. Thanks for the "it isn't 45 people" hint (unless i'm utterly wrong ofcourse). Edit: Alec thought the same lol...

@cooledcannon 10 месяцев назад

finally one i got

@aspidoscelis 8 месяцев назад

Probably by inhaling and drinking, respectively.

@terrysbookandbiblereviews 10 месяцев назад

Trash is great for finding artifacts in Archeology.

@MilitantPacifista 9 месяцев назад

Wait what? Folks in NY pee in bottles? Call me old school, I'll just find the next tree and go at it.

@Hurricaaane 10 месяцев назад

I'm sorry to ask but, is there a place somewhere (description, page...) where I can find out which episode matches a RU-vid highlight? There's times I see a highlight and want to listen to the entire episode, but I can't find a summary of the list of guests for all episodes on the Lateral website. I'm confused whether there's a way to find out that I missed out. Thank you in advance

@marloestwigt5299 10 месяцев назад

You can tell by the date of the highlights/ episodes. The highlights are posted in the week following the episode.

@Poldovico 10 месяцев назад

Here I was thinking a sample of 45 studies in their meta-analysis

@zacharyxx0 10 месяцев назад

Where can I watch the full thing?

@almerindaromeira8352 10 месяцев назад

Every year I see news about the rising consumption of drugs in european cities and all those articles mention analysis of the sewers

@rickys7883 10 месяцев назад

Yes exactly! They used it to track coke and stuff.

@MeriaDuck 10 месяцев назад

Judging from one of the other questions, Hevesh5 deserved to be on your invitation list one day 😂

@lateralcast 10 месяцев назад

Good suggestion! I'll reach out.

@nathangamble125 10 месяцев назад

My guess is sampling litter or landfill sites.

@nathangamble125 10 месяцев назад

Oh my god fsdfdg, sewage was actually my first thought, but then I thought it wouldn't be representative because nicotine and alcohol get metabolised and because most bottles and cigarette butts don't end up in sewage.

@mickymoist 10 месяцев назад

Alec! Is this his/your only appearance on a Tom Scott project? If not, I need directions you the others... lol

@loppydisk 10 месяцев назад

I knew it right away because of the covid testing

@mbdtfnl 10 месяцев назад

In 2012, they measured that in Volendam, NL, the small town consumed the most cocaine per citizen, of any city/town in Europe, even more than big cities like Paris, London or Milan😀 (it was 3rd in Europe when it came down to XTC). This was also measured by looking at the wastewater.

@umey3445 10 месяцев назад

because their confidence interval is massive or something lol

@TheVoidSinger 10 месяцев назад

Knew this too soon, thanks to covid wastewater testing (and learned about all the other things you can test for). Oh and Storm water runoff testing is a thing too.

@jd7863 10 месяцев назад

I really thought Alec had it! Guess I got to be the smug / angry screen shouting person today.

@bigsarge2085 10 месяцев назад

👍👍

@privatesolofoe 10 месяцев назад

Huh, I guessed this one right but I really wasn't confident in it cause I wasn't sure if that's how sewage wastewater worked at all lol

@kaishang6406 11 дней назад

I guessed the same as alex.

@lauxmyth 10 месяцев назад

I wonder what were the error bars for this sample.

@lateralcast 10 месяцев назад

In truth, pretty wide! www.sciencedirect.com/science/article/pii/S0160412019340784

@lauxmyth 10 месяцев назад

@@lateralcast Thanks. I have seen higher relative error for some astronomy measurements. But what are a few thousand light-years between friends.

@Vousie 10 месяцев назад

So then how did they get the numbers so skewed? 45 is not such a small number if it's 45 sewage treatment facilities - that would be a large proportion of the entire country. Or are you saying that you think the numbers are accurate and it really does average out to 3 cigarettes per person per day?

@lateralcast 10 месяцев назад

The point is that the numbers AREN'T skewed - in fact, they lie within the expected range. The study had a large margin of error, but it correlated to the figures generated by more 'conventional' survey methods, such as censuses. www.sciencedirect.com/science/article/pii/S0160412019340784

@57thorns 10 месяцев назад

Not really a large portion of the country. Wikipedia claims that just in Beijing they build one sewage treament plant "since 2000", but it seems that data ended in 2008, but still, that would be at least ten (with a very large new one built fairly recently). en.wikipedia.org/wiki/Water_management_in_Beijing Anyway, wastewater treatment plants are not that few, with at least one for every town or city in countries with decent water treament.

@Vousie 10 месяцев назад

@@lateralcast Ok. Guess it was just the way you guys emphasised the tiny sample size and guessed it was being skewed by extremes that I thought you guys were saying from the start that that study is unlikely to be accurate.

@VonOzbourne 10 месяцев назад

Admittedly Jacklyn was a bit hung up on the skewing that comes from small sample sizes, but this kind of ignored what the samples were representing and how averages work. Granted statistics are useless until you place them in context anyway, but this kind of general baseline is a decent enough starting point.

@aclonymous 10 месяцев назад

I was about to make a comment about how Alec was so confident at the start and ended up being completely wrong, but it turns out my confident guess at the start was also completely wrong. I thought that the sample was 45 air quality readings across China and the conclusion was that the average Chinese citizen breathes enough smog to do the same amount of harm to their respiratory system as 3.1 cigarettes (and ditto with motor function and 8.1ml of alcohol).

@TalasDD 3 месяца назад

45 wastewater treatment plants

@tonypang83 10 месяцев назад

Thought the cigarette consumption would be higher to be honest - the mental image (stereotype?) I have is of the population counting how many packs a day they smoke, not how many individual cigarettes 😅

@jamesirvein5458 10 месяцев назад

In this particular case, you're averaging daily nicotine... uh, output? 😉 across the *entire* population, including children. That should (hopefully?) bring down the average compared to calculating the average adult's consumption...

@ndulmillenium 10 месяцев назад

Oh, yes! As a regular viewer of Beryl and Tom, this is the best crossover ever, after seeing Geoff Marshall on 'ere, what's going on?!

@hannahranga 10 месяцев назад

I'm curious how they decided it was the average adult numbers given there's no way to tell how much booze/nicotine was used by minors

@boy638 10 месяцев назад

Jokes on the Chinese gov, I do my business on the streets to prevent them from tracking my consumption habits.

@uplink-on-yt 10 месяцев назад

I thought Alec was more into brown patterns, not flat blacks...

@MrNathansamson 8 месяцев назад

3.1 is not that weird of an average number right? if 1/10th of the population smokes, and they smoke one package (lets say 30 cigarettes) a day this would be it. I dont know about china, but does not seem far fetched to me?

@Tomtit_Andy 26 дней назад

I would doubt accuracy of measurement based on the waste water in China. China is a big country with a huge diversity of the development across the land. Significant chunk of Chinese population live in villages and underdeveloped towns, where central suige system isn't a thing (to my knowledge). So I guess this measurement is accurate for the main cities of China, but not for the whole country.