Various attempts at vegan analytics

[NonZeroSum]

Video

Veganism Declined by 50% in Two Years: A Warning for the Movement.

VEGANISM IS IN DECLINE (Eisel Mazard response)

Vegan Footsoldier is Incorrect: Short-Term Decline of Veganism, the Stats.


Monthly Views Difference

Philosophical Vegan Wiki - Vegan YouTubers | Analytics | Monthly Views Difference







Subscribership & Average Views Difference

Philosophical Vegan Wiki - Vegan YouTubers | Analytics | Subscribership & Average Views Difference









Conversation

@footsoldier @Theo You realize that if you use "views per video" you'll also be using non-synchronous data, non-ordinal data? You'll be ranking non-ordinal data: the value of one unit won't even be the same for two different points in time (the before and after) for the same channel. https://en.wikipedia.org/wiki/Level_of_measurement

Total number of views/uploads = average views per video, I understand that success with earlier videos can appear to bolster the number against newer ones not doing so well, I think that's what foosoldier was getting at with wanting to be able to measure average video views on new videos/number of new videos alone each month.
It's definitely not perfect.

@Theo There are a lot of depressing "implicit" social factors shown in this list (of vegan youtubers)… like… relationship between gender and political content… i.e., the successful youtubers are predominantly female (but with "lightweight" content, to use a flattering term)… but the "serious" (political youtubers, or even serious HEALTH youtubers) are predominantly male…
Seriously… looking at a lot of cooking in bikinis at the top of this list…
Cooking in revealing attire + workout videos.

@a-bas-le-ciel (Youtube name) if veganism were a financial market and you were trading, would you short sell?
if so you would be trading against the trend at a support level
which means you have a low possibility of being correct and infinite potential to lose money
basically you would lose money as a trader if this is your view
if you think I am dishonest, then fine, but please dont take up trading. sounds like you'd suck

Tis interesting. Another hard thing to measure would be whether viewership is just more dispersed than in the old youtube days, i.e. that there are a lot more vegan youtube channels overall, newer smaller ones that are taking up the slack of the older bigger ones people watch less of now.

@Theo IMHO, no: the smaller channels have suffered even more (i.e., just based on 5 years of observational data… but I have watched the progress of a HUGE number of small vegan channels during that time…)
@Theo The main skew (i.e., bias) we have creeping into this study is the fact that we're excluding entirely channels that have completely collapsed and disappeared —and, also, channels that gave up on being vegan. Somebody like Jason Pizzino is an example of both, TBH.
If you included all the people like Jason PIzzino & Joe Best… the numbers would continue to decline.
Powsimian is under 200 views per video, Dom Bower is under 100 views per video… man, the small channels GOT SMALLER…

and we're not counting the small channels that got smaller, like Vegan Revolution, etc. etc.

Man, it is only an opinion, but it's a pretty well informed opinion: I think that if you extended the study to include those smaller and smaller channels, the numbers would look worse and worse
@Theo The viewers that small channels could get, in the old days (myself included!) were all part of the ecosystem of those 12 million regular viewers Freelee brought in: I would get 40,000 viewers on a video criticizing Freelee… that whole ecosystem is now gone.

It's just heading away from monopoly. There are more overall views but more channels recieving those views with less views per channel
The change in algorithm confirms this
At one point vegan YouTube consisted of durianrider
Now it is extremely diverse
Eisel you conveniently ignored Theo's chart of all channels which clearly shows this
It's still possible for someone to become popular due to genuine interest but the YouTube algorithm now spreads out user attention to hold attention longer
YouTube wants you to watch longer
And watching a more diverse selection of videos holds that attention longer
If you exhaust your sub feed and then turn off the computer, YouTube loses
This is why they have been trying for ages to kill the sub feed
That is completely uncontroversial

[teo123]

My guess is that much of it has to do with the new recommendation algorithm by YouTube, which is allegedly biased towards the left-wing US media. Anarcho-capitalists on YouTube often complain about it.

[NonZeroSum]

Rough no.'s for what can be seen looking at vegan YouTube today:

Active vegan channels: 1493
Vegan channels taking a break or quit making videos: 248
Vegans who deleted all their videos: 31
Vegans who deleted their channel: 17
Ex-vegans: 36
Unsorted: 272

Vegan YouTubers

Vegan YouTube Analytics

Excel Document for the above two.

As well as audience comparison with other campaign movements on youtube over time, it would be interesting to know what content within veganism is on the upward or downward trend.

If anyone wants to help by copying tables over to their own google spreadsheet, filling in content type per creator, new names, ect. Then linking here and we can combine them.

I'll reach out to faunalytics on twitter when I get August results in, see if they're interested in the data.

As well as controlling for YouTube's algorithm/curated content system driving people to more commercial channels, so less small user creator driven, you'd have to figure out the average number of drop offs over time per subscriber success rate with any content, for people simply moving onto other hobbies.

[NonZeroSum]

August results are in:

Vegan YouTube Analytics & Google Spreadsheet.

[brimstoneSalad]

Did you reach out to faunalytics? It'd be interesting to read some interpretations of this data.

[NonZeroSum]

Did you reach out to faunalytics? It'd be interesting to read some interpretations of this data.

Aye they have a call in session I'm going to join tomorrow at 12:00 EST and I sent them some info ahead of time. Anyone's welcome to join and will let everyone know how it goes after also.

What other content consumer demographic on youtube do people think would be good to use as a control group? I'd be happiest getting stats on leftist political channels or more specifically democratic socialists because it's the other content I watch most of and helps me build my video library, but I do think it has a lot of similarities for small creators putting a lot of effort into educational advocacy, less grifters like you see with the anti-sjws, plus a lot of street activist footage. Or could go for something like LGBT content, history, science, etc. Let me know your suggestions.

[brimstoneSalad]

That's a tricky one: I'd say do something that will act as a good control.
I'm not sure that political issues would be, since it's hard to evaluate. Maybe you can do something commercial like a product or trend you can get hard numbers on in the market.

[NonZeroSum]

@brimstoneSalad can you verify Keegan on the wiki sil vous plait.

Things are moving along well, future stats are set to autogenerate, we're implementing mechanisms so that margin of error will be minute, we'll be able to import a sample of historic socialblade stats as far back as they go and we're creating detailed labels to create varied tables out of.

Vegan YouTube Directory - Ideal future format

Vegan YouTube Directory - Data Labelling Instruction Manual

Vegan Youtube Directory Report 1st Draft

Developers notes

[NonZeroSum]

Created an ideal future finished format for a wiki page on Vegan YouTube Analytics

The main goal being to create deatiled monthly analytics on total viewership interest in veganism over time, as well as difference between various vegan sub-topics and a sampling of analytics on new videos per month, carefully accounting for anomalous situations.

So what are some potential problems we can decide how to control for?

Main question: what other youtube communities do you think would make for a good comparison control group? It can be anything from hunting channels to unboxing videos, but preferably have some similarities like the lgbt community for being an identity and an activist movement.

[brimstoneSalad]

Main question: what other youtube communities do you think would make for a good comparison control group? It can be anything from hunting channels to unboxing videos, but preferably have some similarities like the lgbt community for being an identity and an activist movement.

Maybe unboxing videos of particular products, but you'd need to do it on a company that's publicly traded where you can get hard stats on their sales so we have some idea how youtube translates to the real world in economic terms.

Or... if we're talking food, we can look at concrete trends by country and newer food ingredients that we could find import/export stats on.
Find recipe videos featuring the newest superfood, and then look at sales stats on that... might be harder to find than for a specific product (like a toy) though.