Blog analysis

Nope, this isn’t more advance flogging for The Liblog Landscape 2007-2008. (You’ll get that soon enough, along with a special offer for early purchasers. If you’re wondering: I uploaded the PDFs to Lulu yesterday, and am now waiting for the proof copy, which could take a couple of weeks.)

This is a Friday funny–and a slightly delayed joining in an offhand meme I saw at Helene Blowers’ Library Bytes. Namely, a few blog analyzers…and in this case, how they rate this here blog.

Typealyzer

This one claims to do a Myers-Briggs style analysis of the blog. I just ran it and came up with:

ESTP – The Doers

The active and playful type. They are especially attuned to people and things around them and often full of energy, talking, joking and engaging in physical out-door activities.

The Doers are happiest with action-filled work which craves their full attention and focus. They might be very impulsive and more keen on starting something new than following it through. They might have a problem with sitting still or remaining inactive for any period of time.

Bwahahah… yep, that’s me, out there playing volleyball when I’m not on the links or joking with my huge array of friends, since I’m so attuned to people. And, you know, never following through on anything, which is why Cites & Insights disappeared after six issues and I’ve never managed to complete any of those books I’ve started writing…

What’s most absurd here is that I ran the same site’s test a few days ago (November 10), when Blowers posted her item–and came out INTP, The Thinkers (and an introvert). I’m notoriously an introvert (and yes, I tested that way on a real Myers-Briggs test–I think it was INTP. That was when I was LITA Vice President/President-Elect; the LITA Executive Director had all of the Executive Committee members take the M-B test and assured us we were all “E”s because, after all, how else could you win an election?)

So in four days with, I think, two posts, I’ve gone from introvert to extrovert and from one who regularly finishes projects to one who leaps from idea to idea? Man, those must have been some impressive posts…or this is an unusually silly beta site.

Genderanalyzer

This one “uses Artificial Intelligence” to determine whether a blog is written by a man or a woman.

The robots say I write like a man (75%, whatever that might mean).

Well, at least that’s not as absurd as ESTP; last I checked, the gender choice was right.

Readability Test

What grade level this blog is written at.

Junior high school. Whereas C&I is at college/undergrad level and my personal website (which includes a few old articles) is at “Genius” level.

Hmm. I’m happy enough with “Junior high school” for the blog, but I wonder whether that’s as erratic a rating as the Typealyzer. Somehow, though, I find it hard to believe that the handful of items on my personal website are that much deeper intellectually or confounding in style than C&I–or, for that matter, that C&I is all that much more hifalutin’ than this here blog.

In this case, it’s a repeat performance: I posted about this site last November. Came out junior high school then, too–and most individual essays from C&I came out high school, which was fine with me.

What is your blog worth

Claims to determine the blog’s monetary value based on Technorati ranking and advertising potential.

$28,227.00

I love the precision: Not just $28,000, but $28,227.

Ah, but the “most successful linkbaiter, ever” who put up this site claims his blog is worth $6,220,101.72. Must be nice to be rich. (And yes, that site has LOADS of ads. When I had ads here, I made $24–over six months, as I remember. Yes, my Technorati “authority” rating and number of subscribers have both grown. Not that much, though…)

As Helene says, “use at your own risk” and “for your pure amusement purposes only.” By the way, she comes out ISTJ, woman, high school and $47,569.

Oh, and if someone wants to sponsor this blog, I’d ask a whole lot less than $28,227 (but more than $227)–but then, is that “per year” or “over a lifetime”?

In fact, I am very much looking for sponsorship–but primarily for ongoing real-world research into blogs and wikis, with this blog being a tertiary possibility.

Meanwhile, you may find these tests fun. If you test at the genius level…well, I probably couldn’t understand all them big words anyway.

16 Responses to “Blog analysis”

  1. Dorothea Salo Says:

    CavLec came out as INTP, which is actually pretty close to the I(N/S)TJ I usually come out with on M-B testing.

    I’m really curious about their algorithm… which is soooooooo the NT talking.

  2. walt Says:

    Isn’t it though? There was a brief LSW conversation about algorithms… (LSW is about as E as I get.)

    My own two-sample experience says the algorithm is easily gamed or unusually time-sensitive. Or something. Maybe it’s a random letter generator?

  3. John Dupuis Says:

    INTP, which is what I usually score on the MB test itself. We should start a club.

    And it would be interesting to regenerate the results every few blog posts over a longer period of time and see what the variation is.

  4. Dorothea Salo Says:

    I’m up for the experiment if y’all are!

  5. Mark Says:

    If I remember correctly, I came up INTP or something quite close, gender neutral, and junior high school, and $8xxx.

    I’m not posting enough to participate in any experiments, unless you just want the randomness of the algorithms checked.

  6. walt Says:

    Interesting notion.

    Oddly enough, when I was checking the 2008 metrics for The Book, Dorothea, John and I all had nearly identical posting frequencies–I had 68 posts during the three months, Dorothea had 65, John had 64. (Trust me: That’s a VERY tight cluster–out of 533 blogs with countable posts for that period, one other has 68, three others have 65, two others have 64–and NONE have 66 or 67 posts.)

    I don’t think any of us are posting that often any more, but still…

    OK, here’s a proposal: Check your Typealyzer results either twice a month or after every 10 posts (whichever takes longer–this spring, those would have been about the same at roughly five posts per week) from December 1 until, what, February 1? That would yield six samples (including the one you’ve already done). Then send me the results along with any comments about what might have affected Typealyzer; I’ll put together a post (or maybe a C&I piece?)

    Too short a sample period? Keep going. Too arduous? Don’t do it. Anyone else want to join in and fairly certain that participating won’t affect your blogging style (thus gaming the results)? Please do.

    Introverted thinkers of the world, unite–well, no, that doesn’t sound right.

  7. Wayne Says:

    This is eerie. Mine came out INTP as well, which is also my M-B type. It’s at a high school level, but unfortunately only worth $14,678.04. I would have guessed $14,678.12, so that’s pretty close to my own calculation.

  8. Mattias Östmar Says:

    Dorothea Salo: I´ve spent an awful lot of time (INTP/INTJ myself…) training a simple classifyer with examples based on books and Jon KÃ¥gström (INTJ) have developed his classifier over and over again to get these results. I thinki it´s wise not to over-interpret the results from the Typealyzer, since, as Walt says, it needs much data over time to se the “normal” type preferences. And I personally believe that people might use their blogs as a way of extraverting a certain persona rather than how “they” are. But I´m very exited about the results you guys get!! :-)

    @walt: maybe we could provide you with an easier and quicker way to do such tests (we would love to get help to evaluate and improve the system) – we´re thinking about an API (needs som programming skills), a box for you to put in your RSS-feed to automatically analyze and track your own or others blog posts and we´re also developing a reserach tool wich uses blog aggregators such as http://www.twingly.com to let people and companies analyse a lot of posts at the same time. What would be helpful for you?

  9. walt Says:

    Mattias: If I do a “small sample over time” thing–which depends on whether John and Dorothea send reports–it will be just that: An amusing anecdote.

    I really wasn’t planning to add this facet to the liblog analysis I do already: it’s just not central to anything else I’m doing. If I hadn’t been so startled by what I regard as an implausible move from Introvert (which I am) to Extrovert (which I’m not) based on the addition of three or four posts, I wouldn’t have commented as I did.

    Maybe someone else would take this up as an actual study; I already have more potential things than I have time or energy for. Sorry.

  10. John Dupuis Says:

    I checked again today and it’s still INTP. I’ll be putting weekly tests at the end of my Friday Fun posts for the next few weeks to see what happens. Btw, Wayne, the AL is ISTP so I guess he/she/they can’t be you ;-)

  11. walt Says:

    Hmm. Still ESTP. Guess I need to write some deeply philosophical, antisocial, introverted posts. But that would be gaming…

    So maybe I have a blog persona that’s [becoming] more outgoing than my real-world persona? Possible.

  12. Kathryn Greenhill Says:

    Um… Walt…it’s really, really embarrassing to admit, but my blog comes out as worth the same bizarre value as yours… $28227 (but I have to use the URL http://librariansmatter.com/blog for it to pick up anything). You’re the guy with the numbers, can you pick up any reason *why* we would come out so similar???

    BTW – According to Gendalyser I’m also 81% male . I’ve tested the same on an analysis of my twitter stream too.

  13. walt Says:

    Kathryn: The kind of numbers behind that value rating are way beyond me. Based on Google Page Rank and Technorati, your blog should be worth more: Same GPR (5), but you have a Technorati authority of 96 (which should mean 96 blogs have linked to yours in the last 90 days) while mine is a lowly 50.

    On the other hand, popuri.us shows me as having more Google links and Yahoo links. (The latter figure may be meaningful; the former almost certainly isn’t.)

    The site says it uses Technorati’s API. You should show more value. So, no, I have no idea what’s going on.

  14. Kathryn Greenhill Says:

    Ah… maybe it has a “I don’t want boy germs” function built into the algorithm and my obvious maleness makes me slip further down the scale :)

    As an aside, I get annoyed with Technorati rankings because it tends not to account for repeat links from blogs. So, none of Michael Stephen’s links to my blog have shown up for over a year, although anyone who links to me who never has done so before is counted. With the small number of llibblogs out there, one would think that after a couple of years there were no new libblogs to link to me so my T ranking would plummet…

  15. walt Says:

    As I understand it (which ain’t a whole lot), Technorati’s “authority” represents the number of blogs with links in the past six months (I keep saying “90 days,” which is wrong), while the link count is the number of posts within the past 90 days–and, I think, blogrolls can influence both counts, depending on how the results are gathered (now there’s a handwave for you). (At least I know I see links where my blog only shows up in the blogroll.)

    But that’s just my understanding, and may not have much to do with what’s actually going on.

  16. Joerg Kurt Wegner Says:

    @John: I took the genderanalyzer test and now I know that I am to 60% a woman. Mmmhhh, I got to talk to my wife about this.


This blog is protected by dr Dave\\\\\\\'s Spam Karma 2: 103753 Spams eaten and counting...