73 votes

The 2019 Tildes Census: For real this time

45 comments

  1. [18]
    Grzmot
    (edited )
    Link
    Hello everyone, again. Hope I'm not annoying yet. I've worked on the form based on feedback from the last two threads on this subject. I know it's not entirely perfect yet, but I don't think think...

    Hello everyone, again. Hope I'm not annoying yet.

    I've worked on the form based on feedback from the last two threads on this subject. I know it's not entirely perfect yet, but I don't think think it's realistic to make the form suit everyone. Someone will always be in a life situation which just isn't going to fit a box in a form. I hope that in this case the Other textboxes will fit your needs.

    In case you haven't followed this debate, a short FAQ:

    Why are we doing this again?

    Because the first form was garbage and the site was too. After the first 70 submissions I couldn't access any of the results anymore. JotForms is a commercial, GDPR compliant provider of forms which I think fits the bill for everyone. They're reliable when it comes to traffic much larger than Tildes could ever achieve today, and they won't lose my data. For the privacy-conscious folks: They also don't give away any of the collected data, because they aren't the controller, I am. Be assured that I will not give the collected data away to anyone and will use it solely for creating fancy graphs and to provide the Tildes community with demographic information. Your submissions will only be released in an aggregated, anonymized format (i.e in statistics and graphs) and I don't plan on releasing individual submissions at all. Should such thing be required, I'll ask the person who submitted for permission first (obviously).

    What if the provided options don't fit me?

    I tried to include everyone's criticisms to certain degree. It's not possible to make everyone happy. Most options have an Other box in which you can enter your specific answer if the provided ones don't fit at all. I would kindly ask you to check the closest matching option if it fits at all. I assume this conversation will come up mostly due to the education and work related questions. The fields of work are general groups. If you follow the link related to that question you can click on each group and see what exactly is in there and if that group applies to you. I think that's the best way to do things. When it comes to education, the different ways education works around the globe make it hard to generalize. I've included options I didn't even know existed, so I definitely learned something while making this form.

    What if I have suggestions to make the form better?

    Please, be my guest in the comments. However do note that I won't be repeating this experiment for the 2019 census. I plan to make a 2020 census, and will use my new experience and knowledge then, but I'll also create a preparation thread here on Tildes well ahead, maybe in November/October 2020 so everyone can chime in. Expect the next census in December 2020.

    How can we thank you for this divine service, great one?

    Glad you asked, though I don't know why you're wearing cultist robes. I purchased the lowerst tier of the Premium tier on JotForms, as the 100 submissions limit would probably have been too low for Tildes (The last census got around 200 submissions, I think). I don't want to make a profit from this, but it'd be nice if I could at least cover my costs: https://www.paypal.me/GrzmotTildes

    If you wish to donate with other means because you don't use Paypal, send me a PM, I'm sure we'll find a way.

    Not donating is also fine, and I'm happy to be supporting this great little community.

    Have a nice New Year everyone.

    ADDENDUM: 37 41 submission in 4 hours, great going so far. :) I presume it's going to get much better once the Americans wake up.

    30 votes
    1. [17]
      hungariantoast
      (edited )
      Link Parent
      It is too late for this year and it is my fault for not bringing it up sooner, but do you think next year you could actually release the new data for 2020? I know I would love to play around with...

      Be assured that I will not give the collected data away to anyone and will use it solely for creating fancy graphs and to provide the Tildes community with demographic information. Your submissions will only be released in an aggregated, anonymized format (i.e in statistics and graphs) and I don't plan on releasing individual submissions at all.

      It is too late for this year and it is my fault for not bringing it up sooner, but do you think next year you could actually release the new data for 2020? I know I would love to play around with it myself and I am sure others would too.

      Beyond that, and I don't mean any offense by this, since I am sure your graph-crafting skills far exceed my own, releasing the data and allowing more than just one user to use it in creative ways is almost surely going to produce better results as Tildes continues to grow, wouldn't it?

      7 votes
      1. Grzmot
        Link Parent
        It definitely would, but I'd have to add a disclaimer to the form and now it's too late. It's definitely something to add for 2020 though, good point. You're right, the more hands are on it, I...

        It definitely would, but I'd have to add a disclaimer to the form and now it's too late. It's definitely something to add for 2020 though, good point. You're right, the more hands are on it, I think the more creative people can get.

        The best thing I can offer you now is to fulfill specific requests.

        11 votes
      2. [15]
        Kuromantis
        Link Parent
        Because of privacy. A lot of people here don't want their data given out for any reason without being reassured it's not going to be abused, even if solely for statistical purposes.

        It is too late for this year and it is my fault for not bringing it up sooner, but do you think next year you could actually release the data? I know I would love to play around with it myself and I am sure others would too.

        Because of privacy. A lot of people here don't want their data given out for any reason without being reassured it's not going to be abused, even if solely for statistical purposes.

        6 votes
        1. [10]
          hungariantoast
          Link Parent
          Are the answers to these questions really that sensitive? I like to think that I'm a relatively privacy-conscious person, but there isn't anything in the content of that survey, considering that...

          Are the answers to these questions really that sensitive?

          I like to think that I'm a relatively privacy-conscious person, but there isn't anything in the content of that survey, considering that the answer data is anonymized and detached from this username, that I would not be okay with being released publicly.

          But, like I said, this year's data has already been promised to not be released. So, what I am asking is, can we, next year, agree to release the 2020 census data publicly?

          7 votes
          1. [4]
            VoidOutput
            Link Parent
            Or at least add a mandatory yes/no question at the end asking for consent.

            Or at least add a mandatory yes/no question at the end asking for consent.

            5 votes
            1. [3]
              hungariantoast
              (edited )
              Link Parent
              I feel like that might negatively affect the data. Of course, planning to release the data might also negatively affect the data by causing some users to just not take the survey at all. However,...

              I feel like that might negatively affect the data. Of course, planning to release the data might also negatively affect the data by causing some users to just not take the survey at all.

              However, I feel like not having a consent question will lead to better, more accurate, and more truly representative data than providing an opt-out option would.

              I also feel like the number of users that just wouldn't take the survey because they know the data is going to be released would be much smaller than the number of users who would take it, but then opt-out at the end.

              4 votes
              1. [2]
                VoidOutput
                Link Parent
                I meant for this question to affect your answers only, it wouldn't be a vote. That way, graphs depicting average data could still be drawn, but if raw data were to come out, your data would only...

                I meant for this question to affect your answers only, it wouldn't be a vote. That way, graphs depicting average data could still be drawn, but if raw data were to come out, your data would only be included if you consented. Sorry if that was already clear, it feels like there was a misunderstanding.

                3 votes
                1. hungariantoast
                  Link Parent
                  Sorry, I did understand what you meant, but used the wrong words when writing my comment. I replaced "results" with "data" and changed some other bits around. For the sake of clarity, I'm just...

                  Sorry, I did understand what you meant, but used the wrong words when writing my comment. I replaced "results" with "data" and changed some other bits around.

                  For the sake of clarity, I'm just going to rewrite what I originally wanted that comment to say:

                  I feel like giving users the option to opt out of having their answers released in a raw data dump would negatively affect the raw data by reducing the accuracy or representativeness of the data.

                  I do acknowledge that not including an option for users to opt-out from having their answers be released might also affect the data by causing certain users to just not take the survey at all.

                  However, I feel like not including an opt-out option would have an overall less negative effect on the quality, accuracy, and representativeness of the released data.

                  2 votes
          2. [5]
            pallas
            (edited )
            Link Parent
            They are sensitive. If the data are detached from usernames, but individual responses are still fully available, combinations of responses could be used to de-anonymize at least some users (likely...

            They are sensitive. If the data are detached from usernames, but individual responses are still fully available, combinations of responses could be used to de-anonymize at least some users (likely to their personal identity, rather than their username), and some of the questions could have responses that could be sensitive and not publicly known in all contexts. I suspect that around three or four of the responses would uniquely identify me, for example, while other responses are things that I don't publicly disclose when attached to my personal identity.

            However, because of this, I wouldn't submit a response anyway.

            2 votes
            1. [4]
              Weldawadyathink
              Link Parent
              I am trying to figure out what responses might be sensitive. All that I can come up with is possibly being transgender. Everything else seems like fairly open information. Could you please expand...

              I am trying to figure out what responses might be sensitive. All that I can come up with is possibly being transgender. Everything else seems like fairly open information. Could you please expand on which questions you feel might cause issues if someone's answers were known?

              4 votes
              1. [3]
                joplin
                Link Parent
                I would think that options like being non-binary, transgendered, your sexuality, age and religion could all be considered sensitive data in some contexts. Please don't assume that because you...

                I would think that options like being non-binary, transgendered, your sexuality, age and religion could all be considered sensitive data in some contexts. Please don't assume that because you can't see why this could be a problem for some people that it's not a problem for most people. There are many regressive societies today that imprison or kill people based on the above factors. (Well, probably not age, but you get my point.) And in some countries the internet is not open, with all data going through man-in-the-middle proxies that give the government access to everything.

                5 votes
                1. [2]
                  Weldawadyathink
                  Link Parent
                  I do not assume that. I opened a discussion to get more information and opinions rather than ignoring you or flat out rejecting your criticisms. I live in the US, and I do often forget that the...

                  Please don't assume that because you can't see why this could be a problem for some people that it's not a problem for most people.

                  I do not assume that. I opened a discussion to get more information and opinions rather than ignoring you or flat out rejecting your criticisms.

                  There are many regressive societies today that imprison or kill people based on the above factors.

                  I live in the US, and I do often forget that the developed world is not the entire world. Thank you for reminding me. Honestly, I sorta just thought subconsciously that anyone in those types of situations would not be a citizen of the internet or tildes, but I do realize now that this is gatekeeping and not accurate.

                  Are there any other situations besides rogue governments where this information could be sensitive?

                  Also, all of the response options were optional. If I were under threat of prosecution by a survey response, I would not provide that information even if the data were going to be confined to a single person. It just seems to me that the data are either very easy to obtain against an single person in a targeted attack, or that the respondee would not provide the data regardless of the data being released or not.

                  7 votes
                  1. joplin
                    Link Parent
                    I think there are. Things like hate groups, and even just random internet trolls. I can understand that, and I might make a similar decision. However, I can envision a situation where someone...

                    Are there any other situations besides rogue governments where this information could be sensitive?

                    I think there are. Things like hate groups, and even just random internet trolls.

                    If I were under threat of prosecution by a survey response, I would not provide that information even if the data were going to be confined to a single person.

                    I can understand that, and I might make a similar decision. However, I can envision a situation where someone wants to help make the internet or Tildes better for others like themselves and answers because of that, under the assumption that the data will be private. It would be nice to make that possible, but obviously the people who run this site need to make that decision.

                    3 votes
        2. [4]
          AugustusFerdinand
          Link Parent
          Then it seems like the answer to that problem is: Don't provide any data. No one is making anyone answer this census. If your privacy is so important to you that you're unwilling for the data you...

          A lot of people here don't want their data given out for any reason without being reassured it's not going to be abused, even if solely for statistical purposes.

          Then it seems like the answer to that problem is: Don't provide any data.

          No one is making anyone answer this census. If your privacy is so important to you that you're unwilling for the data you provide to actually be used, then why ware you providing it?

          3 votes
          1. [3]
            joplin
            Link Parent
            They're not necessarily unwilling to allow it to be used. They're unwilling to allow it to be released to the general public. They may want for it to be used for the purposes of making this...

            They're not necessarily unwilling to allow it to be used. They're unwilling to allow it to be released to the general public. They may want for it to be used for the purposes of making this website better, but not wish to allow anybody on the internet to download and play with the data. Hence, they want to participate, but don't want to see the data available to the public at large. It's a perfectly reasonable position and your comment is rather dismissive.

            3 votes
            1. [2]
              hungariantoast
              Link Parent
              By submitting answers to the survey, users already are allowing anybody to play with the data. They have no idea who @Grzmot is and they cannot ensure @Grzmot is trustworthy. Now, I'm obviously...

              but not wish to allow anybody on the internet to download and play with the data

              By submitting answers to the survey, users already are allowing anybody to play with the data. They have no idea who @Grzmot is and they cannot ensure @Grzmot is trustworthy.

              Now, I'm obviously not accusing @Grzmot of anything nefarious, I 'trust' them enough to have submitted my own set of answers to their survey, but my point is this:

              Anyone who is so seriously concerned about how the answers and data they provide to this survey might be used against them, shouldn't be filling out the survey at all, regardless of how many people are going to have access to the data.

              And that's not to dismiss those who are concerned about their data, that's a very valid concern, but there's little difference between trusting a stranger on the Internet with the data (who could release it if they wanted to) versus dumping the data publicly. The potential for abuse is the same either way unless you have any way of ensuring @Grzmot can be trusted (which, no offense, I don't see how that is possible via the Internet).

              @pallas, this is as much a reply to @joplin as it is to you.

              This is my take: Next year, the census survey should be conducted and answers provided with the understanding that all answer set data will be publicly released. The "opt-out" option for not having your answer set made public should only be to either not take the survey at all, or simply not fill out questions whose answers you deem too sensitive.

              5 votes
              1. Grzmot
                Link Parent
                Yes, however in the first comment I made here, I explicitly said the following: If I released the data set publicly, I feel like I'd be going against my own words. I've already said that I'll...

                Yes, however in the first comment I made here, I explicitly said the following:

                They [JotForms] also don't give away any of the collected data, because they aren't the controller, I am. Be assured that I will not give the collected data away to anyone and will use it solely for creating fancy graphs and to provide the Tildes community with demographic information. Your submissions will only be released in an aggregated, anonymized format (i.e in statistics and graphs) and I don't plan on releasing individual submissions at all. Should such thing be required, I'll ask the person who submitted for permission first (obviously).

                If I released the data set publicly, I feel like I'd be going against my own words. I've already said that I'll consider releasing the data set next year, but with a disclaimer in the form.

                2 votes
  2. [3]
    Grzmot
    Link
    Hello everyone, so far it's safe to say that it's been going pretty great. As of right now, we've had 227 responses in total, with the most responses happening at the beginning and then...

    Hello everyone, so far it's safe to say that it's been going pretty great. As of right now, we've had 227 responses in total, with the most responses happening at the beginning and then drastically falling off.

    I'd also like to thank everyone for their donations! I've recouped my costs, so any future donation will go to me directly. It's not that much, just a few euros, maybe I'll buy a burger or something. Or donate it to Tildes.

    I'll keep the form open until Thursday, Jan 3rd, unless I get some drastic sudden increase in responses warranting further uptime (unlikely to happen). Once I close it down, I'll start working prettifying the results.Microsoft Excel to the rescue! Also these sup tags really could be better

    8 votes
    1. [2]
      alexandre9099
      Link Parent
      I just submitted my response yesterday :D

      I just submitted my response yesterday :D

      2 votes
  3. [4]
    Flashynuff
    Link
    I'm confused on the 8values section. If I got an 8values result of, say, 87.8% equality on the economic axis, how does that translate into a 1-10 score? Are we using the left side of the values as...

    I'm confused on the 8values section. If I got an 8values result of, say, 87.8% equality on the economic axis, how does that translate into a 1-10 score? Are we using the left side of the values as what we're measuring?

    9 votes
    1. [3]
      Grzmot
      Link Parent
      Sadly I couldn't remove the number scale, so I understand it makes it a little more confusing. Basically you pick on the side you more strongly identify with according to your 8values score. So...

      Sadly I couldn't remove the number scale, so I understand it makes it a little more confusing. Basically you pick on the side you more strongly identify with according to your 8values score. So for example if you have a 87,8% score for Equality, 9 would probably be the best fit for you. If your diplomatic score is something like 10% (strongly favouring Nation), you'd pick 1 on the Diplomacy scale.

      5 votes
      1. [2]
        cwagner
        Link Parent
        wow, for next year you need to dumb that explanation down. I read the description on the site and put my points that way. I didn't even occur to me to take a test on that site. So I guess the...

        wow, for next year you need to dumb that explanation down. I read the description on the site and put my points that way. I didn't even occur to me to take a test on that site. So I guess the lesson is to idiot-proof your surveys :D

        4 votes
        1. Grzmot
          Link Parent
          Hey, the test wasn't mandatory anyway, so there's a good chance a lot of people did it this way. But you can do it afterwards as well, it's a fun test. :)

          Hey, the test wasn't mandatory anyway, so there's a good chance a lot of people did it this way. But you can do it afterwards as well, it's a fun test. :)

          7 votes
  4. [4]
    envy
    Link
    This survey can be submitted multiple times, and can be submitted with blank values.

    This survey can be submitted multiple times, and can be submitted with blank values.

    7 votes
    1. [3]
      Grzmot
      Link Parent
      This was done on purpose, as some people don't want to answer some questions, in particular where they're from and other personal information. I trust that the Tildes community is responsible...

      This was done on purpose, as some people don't want to answer some questions, in particular where they're from and other personal information. I trust that the Tildes community is responsible enough to not abuse this, so far I've only had one joke response and one empty form (my guess is from you).

      11 votes
      1. [2]
        envy
        Link Parent
        The blank form was from me. Sorry.

        The blank form was from me. Sorry.

        3 votes
        1. Grzmot
          Link Parent
          It's alright. No harm done :)

          It's alright. No harm done :)

          5 votes
  5. [8]
    tomf
    Link
    so... in the future, why not just use Google Forms and run all the stats with a query?

    so... in the future, why not just use Google Forms and run all the stats with a query?

    5 votes
    1. [7]
      VoidOutput
      Link Parent
      There were privacy concerns.

      There were privacy concerns.

      21 votes
      1. [2]
        Diff
        Link Parent
        I'm happy choosing to give my demographics to some lil "Tildes-local" dude so he can draw up some graphs because he likes graphs. And I can see some benefit for all of us in that, even if it's...

        I'm happy choosing to give my demographics to some lil "Tildes-local" dude so he can draw up some graphs because he likes graphs. And I can see some benefit for all of us in that, even if it's just satisfying idle curiosity.

        Less happy having my arm twisted into giving up my location, purchase history, travel history, life history, world history up to giant companies so they can try to figure out some way to monetize it.

        11 votes
        1. Grzmot
          Link Parent
          OP here, can confirm that I like graphs. :)

          OP here, can confirm that I like graphs. :)

          7 votes
      2. [4]
        tomf
        Link Parent
        weird. I guess it makes sense.. but I figure the people like me (who are concerned about privacy) wouldn't fill one of these in to begin with. I think GSuite would cover most of the privacy...

        weird. I guess it makes sense.. but I figure the people like me (who are concerned about privacy) wouldn't fill one of these in to begin with.

        I think GSuite would cover most of the privacy issues, since it does not violate HIPAA... but who knows.

        1 vote
        1. Grzmot
          Link Parent
          If that were the case, I don't think they would've complained about Google Forms in the first place. Besides, I think getting away from Google is in general a good idea.

          I guess it makes sense.. but I figure the people like me (who are concerned about privacy) wouldn't fill one of these in to begin with.

          If that were the case, I don't think they would've complained about Google Forms in the first place. Besides, I think getting away from Google is in general a good idea.

          12 votes
        2. HanakoIsBestGirl
          Link Parent
          Even if Google does not look at the actual data entered into the form, there is still a ton of metadata for them to peruse. Plus Google forms require proprietary JavaScript.

          Even if Google does not look at the actual data entered into the form, there is still a ton of metadata for them to peruse. Plus Google forms require proprietary JavaScript.

          9 votes
        3. Keegan
          Link Parent
          If enough people don't fill the form for that reason, the sample wouldn't be representative of all of Tildes.

          I figure the people like me (who are concerned about privacy) wouldn't fill one of these in to begin with

          If enough people don't fill the form for that reason, the sample wouldn't be representative of all of Tildes.

          4 votes
  6. [4]
    krg
    (edited )
    Link
    I do wish a more specific locale than "country" could be picked for residence... The United States is a big country, and what I experience in my parts is probably quite different than what those...

    I do wish a more specific locale than "country" could be picked for residence... The United States is a big country, and what I experience in my parts is probably quite different than what those across the country experience. And I'm willing to bet the same goes for those in other large countries.

    also, i think ethnic and economic background are significant and maybe should be accounted for.

    4 votes
    1. [2]
      mrbig
      Link Parent
      I think the problem with that is that there are lots of big countries out there (and even smaller ones have relevant subdivisions), so the form would also need to take all of those into account.

      I think the problem with that is that there are lots of big countries out there (and even smaller ones have relevant subdivisions), so the form would also need to take all of those into account.

      4 votes
      1. krg
        Link Parent
        oh, for sure, I understand the logistical difficulties. But, still, I think distinctions based on country aren't nuanced enough. Especially as we approach some sort of Internet monoculture ..thing.

        oh, for sure, I understand the logistical difficulties. But, still, I think distinctions based on country aren't nuanced enough.

        Especially as we approach some sort of Internet monoculture ..thing.

        2 votes
    2. Grzmot
      Link Parent
      The problem is that there is no proper easy to use state selection, only a freeform answer. How would state questions about economic/ethnic background?

      The problem is that there is no proper easy to use state selection, only a freeform answer.

      How would state questions about economic/ethnic background?

      3 votes
  7. [2]
    envy
    (edited )
    Link
    In terms of Equality, what is the definition of socialism for the purpose of this survey? Does a higher score for equality mean I am in favor of Chinese style socialism? India style socialism?...

    In terms of Equality, what is the definition of socialism for the purpose of this survey? Does a higher score for equality mean I am in favor of Chinese style socialism? India style socialism? Norway/ UK style of "socialism?"

    Right now I am assuming that 10 means I like Chinese style socialism. Which leads me to the second question. I strongly support progressive tax codes and social programs, but I also strongly support free markets. If I like the balanced achieved in Norway/UK, which number am I? 5? 8?

    Edit: I took the 8values questionnaire, and used the values. I am 69% for equality on the 8values survey, so I scored myself 7 on the tildes survey.

    3 votes
    1. Grzmot
      Link Parent
      It might help to do the 8values questionaire if you're unsure about how to place yourself. It's 70 questions, but it can be done fairly quickly. In regards to your question, it's hard to answer...

      It might help to do the 8values questionaire if you're unsure about how to place yourself. It's 70 questions, but it can be done fairly quickly.

      In regards to your question, it's hard to answer because the 8values value comes from a mix of different statements and how strongly you agree/disagree with them. The values don't refer to any model of socialism directly. With your preference, you'd probably be somewhere around 6-8, but in all honesty, this gauge in particular isn't incredibly accurate anyway.

      7 votes
  8. [2]
    kfwyre
    Link
    Thanks again for setting this up, Grzmot! Have you recouped your costs through donations yet?

    Thanks again for setting this up, Grzmot!

    Have you recouped your costs through donations yet?

    3 votes
    1. Grzmot
      Link Parent
      No problem, I like doing this stuff! :) And no, I got back 9€ and I'd need about 17€ And I have to thank that one particular person who donated 5€, you're currently by far the highest donation!

      No problem, I like doing this stuff! :)

      And no, I got back 9€ and I'd need about 17€

      And I have to thank that one particular person who donated 5€, you're currently by far the highest donation!

      6 votes