Personal Web Repository
When creating Mechanical Turk HITs, it’s necessary to use HTML syntax. This is a list of 20 examples of HTML syntax (tags and attributes) for accomplishing common tasks in Mechanical Turk HITs. These 20 examples don’t cover the entirety of HTML, but they cover all the HTML that is typically needed to setup a Mechanical Turk HIT. A single-page version suitable for printing can be downloaded here.
Continue reading ““Cheat Sheet” of HTML syntax for creating Mechanical Turk HITs”
The Ruby, JavaScript, and Perl files I have posted below may be useful to someone who wants to access MTurk’s APIs to create HITs. Basic tasks (e.g., creating a HIT) and intermediate tasks (giving a “bonus” to a worker) can be accomplished easily by using Amazon’s web interface. Advanced tasks (e.g., sending e-mail notifications to a long list of workers who have completed your previous HITs) cannot be accomplished with the web interface. There is a striking contrast between the ease of doing tasks with the web interface and the difficulty of doing tasks with the API. That difference is akin to the difference between buying a car and building a car entirely from custom parts. Continue reading “Ruby, JavaScript, and Perl code for using Mechanical Turk’s APIs”
I’ve found some limitations in Mechanical Turk’s API for Create Qualification Type with an auto-graded qualification test. As of this writing in 2015, the limitations are undocumented, and they cause the API request to fail for no apparent reason. In this post I describe the undocumented limits. Continue reading “Undocumented limitations in MTurk’s Create Qualification Type”
Every once in a while, I get an e-mail from a MTurk worker who says they experienced a technical glitch with my HIT and they couldn’t submit it successfully. I can usually tell it’s not a scam because they mention things from the HIT itself. (At a purely rational level, it’s probably not worth their time to write long e-mails about a missing $0.59 HIT reward, but such things can become emotional for some workers.)
Continue reading “Make-good HIT for MTurk Glitches”
Like a lot of social scientists, I use Amazon’s Mechanical Turk for recruiting participants for some of my experiments. It’s cheap, fast, and the data quality is about the same as what I’ve seen with student subjects and professional online recruiters. In fact, I use Mechanical Turk for a variety of crowdsourcing tasks, such as data entry and proofreading. It’s an amazingly flexible resource, and I’m a big fan.
However, as of this writing in January 2013, Amazon has recently made a change that I need to post a warning about. In essence, Amazon has changed their requester interface to make “master” workers a secret, opt-out option. If your MTurk project isn’t working well, this might be the reason. Continue reading “Mechanical Turk’s Hidden Option”