Advertising your Experiment & Paying your Participants

If you are active, or you know someone who is active in any kind of online forum, chat channel, or blog community, this is an excellent way to recruit subjects for your studies. Especially, if you're interested in subjects that don't work on language. Be aware that most online communities will be offended or suspicious if you just post in their forum without ever having participated in any of their discussions. Be polite, get to know the people first. Then you can post. Of course, if you have a friend who is already respected by the people in the forum you may take advantage of that (but don't do it to frequently ;-).

Another good or probably the most efficient way of recruiting participants are email lists that you build up over the course of your career (e.g. using software like Doug Rohde's Subjector), or lists of friends, people in your dorm, etc. I think it is fair to say that the more willing people are to help you (a priori, even without any payment) the more attention they will commit to your experiment and the less noisy your data will be.

There are a couple of Lists or websites that recruit participants for experiments. When you use those as a venue to get participants, be aware what kind of population you are attracting for your experiment.

• Language Experiments allows you to post online experiments. They have a decent number of experiments advertised on their site (some for cash, others for a chance to win in a prize draw, and yet others without payment). Obviously, advertising here will result in non-random sampling, since you will get participants who are likely to have participated in a number of language-related studies.
• Linguist List (use FYI cagetory). You can post experiments here that are of interest to the linguistic community, but be aware that most of your subjects will be (at least somewhat) trained linguists (i.e. you will be doing non-random sampling, which may or may not be what you want).
• Mechanical Turk (by Amazon) is serving a database (and searches over it) of H[uman] I[ntelligence] T[ask]s (HITs), which are tasks that can easily be performed by humans but hardly ever by machines. Experiments do qualify as such tasks. You could post an experiment there.

Before you post your experiment anywhere, be sure that you have IRB approval for your study. Read this section on Ethical concerns and IRB approval.

Payment

Online experiments can make payment of subjects a little bit more complicated, but not really. Check with your school what you need to get reimbursed for subject payment. They may require social security numbers from the participants or even signatures, but usually there is some way around that. Below I list some convenient ways to pay/reward participants that do not require you to send out cash or checks via snail mail.

• Prize draws (e.g. the chance to win $100) can attract a large number of participants even if you only have very limited funds. Be aware of the regulations your school and/or the state has on this type of payment. For example, California state law requires prize draws to be open to everyone (i.e. you cannot restrict a prize draw to participants in your experiment). Of course, it's another question how many people are likely to incidentally come by your experiment website that advertises the prize draw (so, you are likely to get a good ratio of participants to people who just enter the prize draw but do not participate in your experiment).
• Electronic Gift certificates (e.g. Amazon, Ebay, etc.) are an easy way to give out small rewards to your participants.
• PayPal allows you to send electronic cash to any person with an email address. It does not require any extra payment (as long as you have a paypal account), and the payee does not need a PayPal account.