Our server bill is due soon and we need your help in paying it. Click the thermometer to learn more.
Note: clicking on the above banners and making ANY purchase returns a commission to Turker Nation.
If you can't see the ad, please click on Shop on Amazon instead. | Want to advertise here? PM Spamgirl to learn more!

Results 1 to 8 of 8

Thread: Would you take a HIT that requires using an application?

  1. #1

    Would you take a HIT that requires using an application?

    Hello Turkers,

    I am designing a research experiment that requires participants to download and use a laptop application for few weeks. The plan is to first post a couple of HITs where I would simply request Turkers to complete a survey and perform some tasks; after that I would require Turkers to download the application. The application is legit and perfectly functioning (it is not a test for a new app); I would tell Turkers a priori the name of the application so that they can search information about it before accepting the HIT; and I would provide them with paid accounts so that they do not have to personally register with their e-mail addresses. My goal is to then monitor the use of the app and survey participants about it.

    I have the following questions for you:
    - How do Turkers feel about participating in a study where they are required to use an application?
    - How do Turkers feel about participating in a study that consists of a series of HITs (one per week, for about 5 weeks) where you will keep being invited to? In other words, the Turkers that accept to participate to the first HIT would be invited back to the following ones and paid for each individual HIT + Bonus at the end of the period.

    I am trying to understand whether it is feasible to run this experiment with Turkers or whether I should recruit a different pool of participants.

    Thank you in advance for any comments and/or suggestions,

  2. #2
    Well, no one's answered your question, so I'll take a stab at it. Honestly, most Turkers are looking for things that are very quick and maximize time for pay, and I can see people being extremely reluctant to install something on their computers from what is, in essence, an entirely unknown source (there are scammers on mTurk). If you were taking work in a setting that was deliberately anonymous, would you install something on your computer? Since this doesn't sound like a web application, that means you're also dealing with differing computers and operating systems. mTurk has no restrictions on hardware or software, so there are going to be people on mobile of various sorts (iOS and Android at least guaranteed), Windows, Mac, Linux -- what does the application require? I'd also be very sure that the software didn't violate any terms of service, since, although Amazon has changed the rules fairly recently to allow software for HITs, people still think software install = TOS violation (again, it's not any more, but reassuring people in some way would save you headaches).

    If you can make people feel they are (1) safe installing the software and don't have to give a third party any information and (2) that the time of setup plus work is worth their while, then you should be OK. If the task looks like an annoyance and the rate is low, people aren't going to do it, and you'll also risk getting review that warn people away from the work. Put serious thought into how much time it takes a new user to set up the application and include that in the rate of the HIT -- please not in a bonus. People don't trust that they will ever see a bonus, so they're not going to spend time up front to install software when there's no guaranteed payoff for that time.

    So, it's feasible to do with Turkers, provided you are able to establish that the study is real, reliable, and worthwhile. I'm not going to say to use another pool of participants -- who shoves work away? You just need to be careful about trust to get people to install the stuff and keep returning for subsequent HITs. (Make the software install task an annoyance and you won't get repeat participants from any pool of participants.)

  3. The Following User Says Thank You to Augusta For This Useful Post:

  4. #3
    Moderator RippedWarrior's Avatar
    Join Date
    Sep 2011
    Thanked 2,453 Times in 1,020 Posts

    Longitudinal or "diary" studies - where the worker completes a series of surveys or tasks over a period time are common on MTurk. The issue for the requester is usually retention - which is mitigated by offering an increasing reward per task, or a bonus at the end for completing all tasks. In general I think workers like this type of task - it's guaranteed work and we all like to have more work.

    I agree with everything Augusta said about downloads. I know workers avoid the inquisit plugin like a plague, but that is because it hijacks your entire computer so you can't perform other functions simultaneously.
    Eschew obfuscation, espouse elucidation. Batch HITs for newbs. 1000 req'd. 5000 req'd.

  5. #4
    Thank you guys! Your comments make perfect sense. I can definitely be very transparent upfront about what the HIT requires; the application has a website that people can check to verify that is actually legit before deciding whether to participate or not. I can also make sure to pay an adequate rate that takes into consideration the downloading time.
    I will give this some more thought and probably ask again if I have other questions.

  6. The Following User Says Thank You to ResearchLab For This Useful Post:

  7. #5
    I can see peoples reluctance to doing a turk that required installing an app, but I would have no problem with it if it was worth my time.

  8. #6
    hi! I am wondering the exact same thing....as I have designed a study that requires participants to download and use a smartphone app. Did you ever launch the study?

    Thank you!

  9. #7
    To be honest I wouldn’t do a HIT where I had to download something to my laptop. I would be concerned that the requester was trying to get some info off my computer. The frequency I would be ok with though.

  10. #8
    Definitely not. There is only one software ever that is ostensibly safe that has been used on Mturk (Inquisit), and it does not appear to work on any computer. Obviously, if you require a software to be installed for longer then the duration of the HIT, then you are doing so for extremely nefarious purposed, collecting data on users and are probably willing to easily comply with any abusive police requests for the government to obtain said data. There is no valid, reasonable purpose possible for which software needs to remain installed on the system for more then 30 minutes during a survey, for which you are not abusing users and stealing private information from them. In addition any valid, legitimate purpose as those done by Inquisit, also do not need any software because they could simply be done by Javascript or some other type of code on the web instead. No one thinks that theft of browsing data is acceptable.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts