Statistics question

alboxer2000

Why was there a peak on July 2019 for new users?

I was just wondering because it seemed like there was an important event that day.

Last edited by alboxer2000 (Nov. 21, 2024 13:24:00)

Za-Chary

Observe that every other July is a valley in the graph.

I think July 2019 aligns with some large amounts of commercial spam that Scratch was experiencing at the time. I don't remember the specific month it happened, but the general timeline seems about right.

GlitchedThrough

7/2019 looks like it was a slight decline on my end. It doesn’t look like a surge began until November, but that was due to COVID

alboxer2000

GlitchedThrough wrote:
7/2019 looks like it was a slight decline on my end. It doesn’t look like a surge began until November, but that was due to COVID

COVID-19 was found on December 2019¹ but was found in the USA in January 2020.²

Last edited by alboxer2000 (Nov. 21, 2024 14:39:13)

GlitchedThrough

alboxer2000 wrote:
GlitchedThrough wrote:
7/2019 looks like it was a slight decline on my end. It doesn’t look like a surge began until November, but that was due to COVID
COVID-19 was found on December 2019¹ but was found in the USA in January 2020.²

I very clearly remember it being known in later November, and heavily reported on in December.

Za-Chary

GlitchedThrough wrote:
I very clearly remember it being known in later November, and heavily reported on in December.

Regardless, I wouldn't expect many Scratch accounts to be made just because the virus was found around then.

What I'm more curious about is why you say 7/2019 was a slight decline when OP's screenshot shows otherwise. Could you send a screenshot of what you see?

8to16

cz3781 wrote:
(#7)
It was summer, in which case, the site was more active, because people didn't have school. And it looks like the site was growing on a trend.

The majority uses Scratch in school tho, not for when they're bored at home (this was technically not true during COVID lockdowns but not anymore)

I can tell because the statistics seem to show a huge downfall of activity during the summer and a huge rise in autumn

To me it seems like a bunch of SEO bots as scratch experienced mass commercial spam at the time, this might be why CAPTCHA is used on the signup form now

Last edited by 8to16 (Nov. 21, 2024 16:19:57)

GlitchedThrough

Za-Chary wrote:
GlitchedThrough wrote:
I very clearly remember it being known in later November, and heavily reported on in December.
Regardless, I wouldn't expect many Scratch accounts to be made just because the virus was found around then.

What I'm more curious about is why you say 7/2019 was a slight decline when OP's screenshot shows otherwise. Could you send a screenshot of what you see?

https://scratch.mit.edu/projects/1100126051/

Za-Chary

GlitchedThrough wrote:
https://scratch.mit.edu/projects/1100126051/

Looks consistent with OP's screenshot: the number of new users (the green line) had a peak at 07/2019.

cz3781 wrote:
It was summer, in which case, the site was more active, because people didn't have school.

In general this is not true; since Scratch is an educational tool, the website is overall more active during the non-summer months. This is supported by the fact that every other July has a valley, not a peak, as I said above.

Lorystories123

I joined scratch in 2019 but it was in november

BigNate469

That's quite interesting, actually.

During July there was an all-time high of over 3 million new users, and the next month there was about a projected average of 500,000.

Analysis of the New Projects graph alongside it shows a major dip in activity at that time.

The New Comments graph reached its annual low at that time as well- with only a few thousand more comments than users.

My guess is a combination of spam and people joining Scratch because of 3.0, and its compatibility with tablets- the few months before and after 2019 (3.0 was released in January 2019) saw a general jump in new users (including me

).

DifferentDance8

Za-Chary wrote:
I think July 2019 aligns with some large amounts of commercial spam that Scratch was experiencing at the time.

BigNate469 wrote:
My guess is a combination of spam

Except commercial spam isn't a big boogeyman in the Scratch community like false reporting is, even I've seen it at least once. So there must be another reason

Za-Chary

DifferentDance8 wrote:
Za-Chary wrote:
I think July 2019 aligns with some large amounts of commercial spam that Scratch was experiencing at the time.
BigNate469 wrote:
My guess is a combination of spam
Except commercial spam isn't a big boogeyman in the Scratch community like false reporting is, even I've seen it at least once. So there must be another reason

As someone who was a Scratch Team member in July 2019, I can assure you that there was a large amount of commercial spam being made on Scratch at that time.

2016s4m29

Its the same this end

https://u.cubeupload.com/2016s4m29/Screenshot8122024141.jpeg

Discuss Scratch