December 12, 2008
"Notes from the Field by Omar Galárraga"
This is a note to thank you for letting us use the hand-held computers from the X-Lab at UC Berkeley. They worked very well: during October-November 2008, when we conducted over 1,700 interviews among men who have sex with men (MSM) and male sex workers (MSW) in Mexico City.
We collected information on willingness to participate and accept economic incentives conditional on being free of sexually transmitted infections (STIs). The target population was young MSM and MSW (ages 18 to 25). The work was completed in collaboration with members of the local community who know the formal and informal meeting places and have worked with the population for many years. The overall objective of the project was to inform future HIV/AIDS prevention projects.
We programmed about 80 questions into the hand-held computers. The interview lasted 30 minutes on average. The participants could respond to the questions directly into the computers or have the questions read to them by the interviewer. The preferred approach was to have the interviewees self-administer the interview because we included questions about their sexual experiences. Respondents tend to be more truthful when the interview is completely anonymous and confidential.
We used standard contingency evaluation methods, including random-starting-point and iterative bidding to estimate willingness to accept conditional cash transfers for HIV and STI prevention. Each subject was asked if they would enroll in the prevention program given a set level of incentives. To avoid a starting point bias in the willingness to accept questions, the computers would pick a random number (from a given range) to start the bidding process: “Would you be willing to participate in the prevention program for an incentive of $X pesos per month?” Follow-up questions would ensue, raising the bid if the respondent did not accept the first offer, or lowering the bid if the respondent did accept the first offer.
The hand-held computers were also very helpful because they were discrete. Some of the interviews were conducted in bars, on busy streets, plazas, and metro stations. Thus, for security and anonymity, to have a very small device was very useful.
Please let me know if you would like additional information about the data collection.
For more information, check out the Xlab web site at xlab.berkeley.edu