Jump to content

Tulpa community Census 2015 Results


Recommended Posts

I dunno if you're holding back because it's not ready or because you wanna be a tease, though.

 

Making graphs and writing text takes surprisingly hugs amounts of time that I rarely have. I guess going with the initial plan of "releasing everything in a month after census was closed" could be better but I just wanted to publish stuff as a go through it.

Link to post
Share on other sites
  • Replies 38
  • Created
  • Last Reply

Top Posters In This Topic

Hey Waffles, thanks for the feedback. We truly appreciate it.

 

The smoothed graphs you didn't like were done that way because we were tired of bar graphs everywhere and on these scales its not like an exact number can tell us anything precise anyway.

 

But I'm not an expert; and we still have some 1-10 and 1-5 scales to go through. If you know a better way to handle them, let us know.

 

---

 

Also, I'd like to clarify some stuff if you allow me:

 

First of all, there's no pretense here, we just noticed that it would be a lot of work to make graphs and a decent write up for a full report, that would take about a month to complete, so instead we split it in 3 parts for a faster delivery.

 

The contest is just an act of generosity from Shinyuu that doubles as an incentive to get more suggestions and the communities slightly more engaged with the project. Its not like the final part is going to be delayed because of it.

 

Regarding being a "guardian of the data": No need to explain why we decided to deliver it only in the part 3. We plan to release it next week, is that ok with you?

 

Edit: I took too long to answer lol I should stop writing and stopping to grab food.

 

Just so you know, several people asked me about the raw data, I told them we planned to release it with part 3 and everybody was cool with that. In fact you're the first person to ask for it.

 

At any rate we really appreciate you wanting to help with this :) thanks

Link to post
Share on other sites

Hey Waffles, thanks for the feedback. We truly appreciate it.

 

The smoothed graphs you didn't like were done that way because we were tired of bar graphs everywhere and on these scales its not like an exact number can tell us anything precise anyway.

 

But I'm not an expert; and we still have some 1-10 and 1-5 scales to go through. If you know a better way to handle them, let us know.

 

Histograms. You're tired of "bar graphs everywhere" because they're usually appropriate for the kind of things you collect. You don't have continuous data, or anything like the resolution or sample size you'd want to approximate it. Besides which, just putting a line between the points severely misrepresents the data in big ways:

  • The line doesn't exist! It's absolutely meaningless, just a curve that's numerically fitted to a very small number of points with little thought to whether that curve is plausible or anything. I guess, I don't know exactly how it's being done here.
  • Categories look larger/smaller than they are, possibly by a factor of 2 if, say, you have a final bin that's much higher than its neighbour. This is basically because the line tries to go through the discrete frequencies, but it looks like the frequency is the area underneath the graph.
  • It makes it look like there are trends/shapes there aren't. You can draw a squiggly line through 5 or 10 points and it will usually look like there's a trend. Especially if your curve looks nice and rounded at the top. But in practice, the trend is like, four high-order polynomials stitched together. I already pointed out one instance of that.

I know lines look cooler. They do, definitely, but they're super wrong. Besides, at n=400 or whatever, the numbers over 5 categories are reasonably significant.

 

 

First of all, there's no pretense here, we just noticed that it would be a lot of work to make graphs and a decent write up for a full report, that would take about a month to complete, so instead we split it in 3 parts for a faster delivery.

 

The contest is just an act of generosity from Shinyuu that doubles as an incentive to get more suggestions and the communities slightly more engaged with the project. Its not like the final part is going to be delayed because of it.

 

Regarding being a "guardian of the data": No need to explain why we decided to deliver it only in the part 3. We plan to release it next week, is that ok with you?

 

I guess I can't complain much about a week now. But, eh, a couple of weeks ago I might have asked you to explain. I mean, raw data is raw data. I can't remember how much personal info there was in there, but it surely wouldn't have taken very long to just post it.

 

That's why I was a little concerned with the pretense. I understand that making all those graphs takes time (I guess). Just that for some, those graphs shouldn't delay actually handing out the data.

Link to post
Share on other sites

I guess we cleared that up, fortunately!

 

Anyone can run a survey and throw a spreadsheet for the interested parties and say "here its done have fun" but that's, well, lame.

 

At any rate, thanks a lot for your observations, I look forward that we'll manage to make the results even better with your help!

Link to post
Share on other sites
  • 3 weeks later...

#The Community Tulpa Census 2015 Results:

 

- Part 1

- Part 2

- Part 3 - after the notes!

- Raw Data

 

---

Personal Notes:

 

There was a huge delay in making the part 3 of the census due to several real life happenings with the team. While we all know this can happen, this does not exempt us from our obligations with the community since we gathered all the data (which belongs to the community) and promised to deliver it in a readable and timely manner.

 

Excuses aside, you'll find below the final writeup and charts of the census, and at the very end, a link to the datasheet containing all the results!

 

Once again, thanks to everybody who participated!

 

PS.: this part was done by the generous member of the the tulpa.info forums Waffles :)

 

---

 

Skills

==================

(and tulpa sex!)

 

---------------------------------

 

This one is all about skills and getting it on inside your head. Oh boy. The first part of each section is just looking through the individual questions, but there are a few correlations noted nearer the bottom.

 

The data here has been stripped of people with no tulpas for obvious reasons. Otherwise, no-one was removed except where mentioned explicitly, or their data was missing for that thing.

 

The correlations quoted here are all r-values - PMCCs. If you're not familiar with this, don't worry. They're values ranging from -1 to +1, where the size of the value indicates how strongly correlated they are, and the sign whether they're positively or negatively correlated. So if I have a correlation of 0.8 between two variables, that means that as one goes up, the other pretty well goes up with it. 0.2 means that they still go up together, but much more loosely. -0.8 would mean that as one goes up, the other goes down.

 

To make these correlations work, the worded options, such as "low", "very high", etc., were replace with numbers, ranked lowest to highest with 0 usually being "No" or "Not at all".

 

Skills

---------------------------------

 

First off, how many people have attempted stuff at all?

 

Charts: http://imgur.com/a/bJFmn

 

Obviously most people have attempted mindvoice. Pretty similar numbers for possession, imposition and parallel processing: about 60% have attempted them, 30% are interested and 10% refuse to even consider the possibility. Switching has fewer takers, only 30%, which is about right if you consider possession a prerequisite for switching, and 60% of possession-attempters also tried imposition. There was no question for visualisation.

 

Next, ability. Remember, these scales are ability, from 0 (not attempted or nonexistent) to 10 (perfect).

 

Charts: http://imgur.com/a/W28Xf

 

92% of people have at least some mindvoice. 66% manage 5 or better. Those numbers are fairly similar for visualisation, although people seem to rank themselves a bit lower there.

 

The numbers are pretty low for the others, though. For possession, imposition and parallel processing the numbers are fairly similar: about 50% have nothing at all and 30% have 5 or better. For switching, 70% have nothing and 15% have 5 or better.

 

Let's screen out the people who didn't even try.

 

Charts: http://imgur.com/a/hJ90K

 

The numbers here are pretty similar for all four. Only 15% still have nothing after trying and 40-50% have better than 5. These numbers are a little worse for imposition than the other three. And it seems like switching isn't much worse than anything else, just that few people tried it, though the sample size is getting quite low for switching here.

 

How long did this take?

 

Charts: http://imgur.com/a/07WIm

 

These questions asked, "How long on average has it taken you?". The first option, "Not attempted or developed yet" is frustratingly ambiguous. Does it mean "I answered 0 for ability"? Or "I'm not done with it yet"? People seem to have gone for the second option, which is a damn shame because a lot of people with ability scores greater than 0 answered "Not developed" for the time, so we lost a lot of potential data for time development. The results for the 'hard' skills - possession, imposition, parallel processing, switching - are pretty sparse as a consequence.

 

Anyway, the mindvoice question is still fine because most people have finished mindvoice. Of the people who have, 70% did it in under 2 months. 7% took more than a year. Similar numbers for visualisation.

 

For the people who judged themselves done with the hard skills, the numbers are fairly similar. 30-40% got done in less than 2 months, 20-30% longer than a year. This time, switching comes off a bit worse than the other three.

 

 

We can look at the correlations between some of these. An obvious one is time vs ability. Removing everyone who ranked their time used as 0, we get no real correlations in mindvoice, visualisation or possession. Imposition, switching and parallel processing each correlate about 0.25 with the time put into them. The sample is kind of small and the correlations are kind of weak, so they're not very significant. Still, that's what you'd expect, more or less.

 

What about abilities between skills? Unsurprisingly, all skills correlate between each other at least 0.2. The best correlated are possession and switching (0.6), imposition and visualisation (0.5), and parallel processing and possession (0.4). No real shockers there.

 

 

Sex

---

 

Oh baby. Alright look away if you're squeamish.

 

First, sexuality. This was covered in the last two parts but here it is again, in one place. In the tulpa counts below, hosts with more than 20 tulpas were excluded.

 

Charts: http://imgur.com/a/THtJy

 

Heterosexuals are hugely underrepresented in the community, host-wise - 47%. People aged 16-24, the kind of age range you're looking at for the community, comes out at like 3% LGBT total. The MLP fandom (this is a good reference point I promise you) is 16% LGBT. And here we are sitting pretty on 53%. Tulpas make you gay, it's confirmed. Only the furry community could possibly out-gay us (they're about 60%).

 

The trend with tulpas here is kinda obvious. They're even less straight than hosts - only 37%. The other categories all grow pretty much proportionately.

 

Well? Do you do it?

 

Charts: http://imgur.com/a/JOtC3

 

47% of hosts have had sex with at least one of their tulpas. About 75% of those are still willing, along with 20% of hosts that haven't done it yet but seemingly are going to eventually.

 

How much?

 

Charts: http://imgur.com/a/SNXi9

 

Looking at the host and average tulpa charts, you can see that tulpas have a lower libido than hosts overall. But looking at the highest tulpa chart, it seems that fortunately a horny host can usually find a willing tulpa.

 

And how much do they enjoy it?

 

Charts: http://imgur.com/a/RSd7K

 

Looks like tulpas enjoy it a bit more. That's not really surprising, given that they should be more sense-immersed than hosts. Still, 50% of hosts rated their tulpa sex as 3 or 4 out of 4, which is pretty good. Maybe you should try it.

 

There are some pretty obvious questions to ask here. How well does tulpa and host enjoyment correlate? 0.7. That's encouraging. Host and average tulpa libidos correlate 0.4. Again, as expected. Just as a sanity check, tulpa count and highest/lowest tulpa libido correlate +0.2/-0.2 respectively - as you'd expect, libido spreads out when you have more tulpas.

 

Libido and enjoyment? All positive and sizeable. Tulpa average libido vs tulpa enjoyment is 0.6, vs host enjoyment is 0.4. Host libido vs tulpa enjoyment is 0.3, vs host enjoyment is 0.4. No surprises there.

 

Surprisingly though, there's no correlation between number of tulpas and enjoyment for either party.

 

 

Mixed correlations

------------------

 

Does sexing your tulpa help your skills? No, no correlation in any of them, considering people who attempted those skills. Oh well.

 

Do better skills increase enjoyment? For the host, only imposition, at 0.2. As you'd expect. For the tulpa, yes, imposition is 0.3, possession and switching are 0.2. The samples for these are quite small, though, so these correlations aren't particularly significant.

 

Kind of disappointing. Tulpa sex doesn't help you force, apparently, though forcing just might help your tulpa sex.

 

---

 

This finishes the Tulpa Census 2015 graphs and writeups!

 

Thanks again to everybody who participated!

 

HERE YOU CAN FIND THE RAW DATA OF THE SURVEY

 

PS.: had to remove some of the formatting. Sorry

Link to post
Share on other sites

 

 

Maybe you should try it.

 

 

Okay, please don't say shit like that in such a controversial matter.

 

Other than that, good job uploading the rest of the data. I'll admit I don't like the text wall format here as on the external site, but nevertheless I still enjoyed reading it.

"Try to get a better understanding of things before making your judgement." -Khan, Metro 2033

 

Link to post
Share on other sites

Seemed like a joke to me, I wouldn't take it seriously. That was, however, my reaction to the "advice" I was given in #tuppers.

Hi! I'm Lumi, host of Reisen, Tewi, Flandre and Lucilyn.

Everyone deserves to love and be loved. It's human nature.

My tulpas and I have a Q&A thread, which was the first (and largest) of its kind. Feel free to ask us stuff.

Link to post
Share on other sites

I don't think it's that controversial. 2/3 of respondents had either done it or wanted to do it. Besides, the results say you're quite likely to have a good time.

Maybe you should try it.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...