You gotta love people who believe the statistics too literally...

[jducoeur]

So I just noticed a comment in gyzki's journal, that he was randomly selected for a poll of ages of LJ users. The randomized request claims,

"The official statistics claim that the most common age for LiveJournal users is 5 and two million (out of 10.5 million users) are under the age of 10."

I found that a tad preposterous, but was surprised to find that the raw stats page does say exactly that. (Very interesting page, BTW, with things like the raw numbers for all of the most popular interests.)

Surely this is someone just believing a computer bug, though. If you look at the raw numbers, they claim that there are nearly a million users each at ages 5 and 6 -- and a far-more-believable 800 at age 7. The numbers for ages 5 and 6 must be some kind of glitch; I'd guess that they are buckets that are used for people who don't actually give their ages, but that's just a guess. If you look at the official stats page instead, it shows the main peak at a more likely 18 years old.

I'm really pretty bemused by the whole tempest in a teapot about the average age of the LJ user. Yes, I do suspect that the stats are off, and read a bit too low, because the older crowd are less likely to accurately state their age. But it's not a worldshaking crisis...

Comments

[rufinia.livejournal.com]

I don't have the year on my birthday. I suspect that not having the year (a lot don't) would skew the raw data *somehow* but why it (maybe) defaults to age six is beyond me.

[cellio]

I didn't list a year when I created my journal, because I wanted my birthday to be available (e.g. on the portal page) to my friends, but I didn't want to publish my birth date to the whole internet alongside my real name. Many older users do this, so we're uncounted in these stats (or incorrectly counted, maybe?).

Just in the last few days LJ fixed this; you can now enter your birth date but control what subset of it is published.

[jducoeur]

See gyzki's theory below. His guess (which sounds extremely plausible) is that if you don't put your birth year in, it interprets that as "0", and translates it as 2000. Which means that anyone who doesn't put their birth year comes through as either 5 or 6 at the moment, depending on their birthday...

[lady-guenievre.livejournal.com]

I wonder whether a lot of people have the "birthday" of their LJ in there instead of their own birthday... isn't LJ about 5-6 years old?

[jducoeur]

Yes, although I'd be surprised if that many people have been on it for that long. My impression (supported by some of those raw stats) is that it was fairly small at first, and grew steadily...

[vairavi.livejournal.com]

I've been on LJ (with another username) since Sep 2000 and already knew people who had had them for a while when I joined. This LJ was created in 2002.
So...there's some stats for you :/

Damn lies, and...

[metahacker.livejournal.com]

"Hmm, oh, this form wants a date. Let me put in today's date..." Well, maybe not.

Lots of people might use 2000 as a fake date, the way I usually use 1900 (except on websites that don't believe people over the age of 70-80 still exist).

Maybe it's a Y2K error.

Maybe there are 1,000,000 journals devoted to people's cats, satellites, and babies.

Maybe their software is broken, and doing subtraction wrong.

Re: Damn lies, and...

[dsrtao]

My fake birthday is 1-1-1970 (we all know I'm a geek) but that skews it the OTHER way...

Re: Damn lies, and...

[metahacker.livejournal.com]

Wow, you too? :)

[corwyn-ap.livejournal.com]

Yup, and New Hampshire has double the per capita consumption of alcohol of the next higher state.

[learnedax.livejournal.com]

Mm, self-reporting...

The current geographic data there is interesting, as well. Sure, California has the most users overall, but per capita Massachusetts is still the highest at 2.7%, compared with CA's 1.4%. Michigan is the only other one close to MA at 2.6%.

Though it's not a very meaningful number, spatial density is also kind of interesting. MA has 22 LJ users per square mile, compared with MI's 4.7 and CA's 3.3.

[jducoeur]

Wow. Hadn't been thinking about the numbers in terms of per capita, but that's actually pretty impressive. 2.7% per capita is a *very* deep penetration, even assuming that that's a mild exaggeration (due to duplicate journals and such) -- not MySpace level, but still pretty remarkable for a tool that is largely flying below the media radar...

[talvinamarich.livejournal.com]

Ah, but how do you define "Media"?

Livejournal is, itself, a form of the "Media". The World-Wide-Web is fast eclipsing Cable in much the same way that Cable eclipsed The Big Three Networks etc.

I almost feel guilty: I read the website for the local paper, and only grab a *paper* copy when I need to cut something out so my wife can turn it in with a Journalism Assignment.

(Wife's a Journalism Major.)

[corwyn-ap.livejournal.com]

MA has 22 LJ users per square mile

I was just looking at the census, there are counties in ME with fewer than 22 PEOPLE per square mile... :-)

[siderea]

Well, if you want to get technical, that's 22 LJ accounts per square mile...

[gyzki.livejournal.com]

So, does anyone think I ought to respond to the survey? I wasn't going to, seeing as: imprimis, the name honeypot1 does not inspire trust; secundus, nothing in the other user's info was very motivating either; and tertius, last I checked, my full birthdate, including year, was up there on my info page.

[umbran.livejournal.com]

The survey seems to come right after a journal entry relating to advertising and such on LJ. If it's not legitimate, they took a bit of thought in setting it up.

And, by what they've said, if you have your full birthdate up, they've taken that information already, so the poll isn't particularly necessary.

[msmemory.livejournal.com]

I think the whole tempest about ages has to do with planned advertising, as much as anything. If they can promote LJ as a venue for 18-25 year olds, then they can sell more targeted ads. If they have to fess up that a serious chunk of their users are married and over 30, they can't sell as many ads.

[ladyariadne.livejournal.com]

And there was/is this big huboo about directing LJ as an avenue for kids/teens versus adults. They are/were planning on making LJ a myspace clone. (I got tired after reading 500 messages intot he 2500 plus on that official LJ thread.)

They appearantly want to send or make "adult" users go over to their sister type board, but I cant remember the name.

The point is that all the adults need to go back and put in their REAL ages as they are using that info to make real plans that could affect adults poorly

[jducoeur]

They are/were planning on making LJ a myspace clone.

Really, I haven't seen any concrete reason to believe this is true. A lot of people are *claiming* that it's true, but it seems to be based on nothing more than supposition. For the time being, I'm attributing the whole thing to mass hysteria, until and unless someone shows that they're planning on making substantive changes. (I mean, good God, look at all the panic about the incredibly *slight* tweaks they're planning to the user info page. Sheesh.)

They appearantly want to send or make "adult" users go over to their sister type board, but I cant remember the name.

"Vox", IIRC. Again, this seems to be a lot of people panicking for no particularly apparent reason. Nothing that the *LJ* staff has said provides much cause for worry, IMO -- it's just a ton of guesswork and assumptions of secret agendas.

I do think that helping them get their statistics right is probably a good thing -- enabling them to make more informed decisions just makes sense. But I'm sticking to my "tempest in a teapot" opinion until someone points me to something they've said that's more sinister than "most of our users are between 18-25", which is just a statement of (probably correct) fact...

[metahacker.livejournal.com]

...because as we know, people over 30 have no money, and never buy things...? Sounds like they might need a different marketing director!

[metageek.livejournal.com]

We buy things, but I suppose we may be harder to advertise to—better impulse control, better ad blocking software.

[cheshyre]

I also know of many people who create LiveJournals for their infants or pets, supposedly from the infant's/pet's POV. Plus all the journals of fictitious characters.

So there's another element skewing the age lower.

[alexx-kay.livejournal.com]

I wonder what the ages on Geoffrey Chaucer's or Samuel Pepys' blogs are :-)

[cheshyre]

Both syndicated feeds, so they don't count...

On the other hand, there was thefirstevil during Buffy Season 7

[baron-elric.livejournal.com]

To further skew the ages, there are pages for pets and infants that give the real age of the animal or baby. These may account for a lot of the numbers under 6.

[gyzki.livejournal.com]

If you look at the raw numbers, they claim that there are nearly a million users each at ages 5 and 6 -- and a far-more-believable 800 at age 7. The numbers for ages 5 and 6 must be some kind of glitch; I'd guess that they are buckets that are used for people who don't actually give their ages, but that's just a guess.

My guess would be, a year-of-birth field with nothing entered is filled with one or more 0s. Later on, the statistical program comes along and reads "00" as 2000. Subtract date of birth from current date, gives ages of 5 and 6 (depending on month).

[jducoeur]

Sure -- very plausible theory. It even explains why the numbers break down more or less evenly between 5 and 6, since we're right at the middle of the year. I suspect you're correct...