Week 2

Week 3

Week 4

Week 5

Week 6

Week 7

Week 8

Week 9

Week 10

Week 1 (May 20-May 26)

Craigslist made it very easy to recruit participants. In fact, within 12 hours, I already heard from 30+ interested participants. The interest was overwhelming; I had to withdraw the ad within 24 hours, when the number of interested participants reached 50. We were only considering a user pool of at most 10 people, so this was a sufficient number from which to start. We also knew that we could always repost the ad on Craigslist.

Jen and I were a little worried that we were offering too much money to our user study participants. Our ad stated that we would pay participants $20 per week to mail in their grocery receipts, complete a questionnaire of their food consumption, and participate in one interview. We then learned from Dr. James Landay, also of the Berkeley HCI group, that the usual payment for user studies is $50 per hour of work. We then concluded that our payment would be fair. 

User Study Lesson 1:  A good payment estimate for participants of user studies is $50/hour.

With all the interest in our project, the next step was to find our ideal candidates through a pre-selection questionnaire (link to pre-selection questionnaire). I re-created the pre-selection questionnaire and e-mailed it out to the interested participants.  Since we didn't know if everyone would have Word or access to attachment, we put the questionnaire within the text of the email.  

The inclusion of an interview to the user study is also something Jen and I added this week. The idea behind interviewing participants, and one of the mantras of Human-Computer Interaction, is the idea that "you are not your custumers." No matter how much software designers may think they think like their users, often we won't know our users' key thoughts and issues unless we personally ask them or watch them use our software. Contextual Inquiry (link to article) is a method of observing and interviewing users where they work and as they work in order to get the best data. After learning about contextual inquiry, I drafted interview questions (link). We also added the possibility of accompanying our users on a routine visit to a grocery store, as part of the concept of Contextual Inquiry. The purpose of this would be to observe how users would use the output of our device - e.g. Would they use a shopping list (as generated by our device)? What are some factors that influence their decision making process?

Interested subjects returned their pre-selection questionnaires to me and I began the process of finding our ideal candidates. Our candidate pool was lacking in the amount of seniors, so I placed a second ad calling only seniors to respond. I received a few more good candidates from this. Our goal was to get a pool in which our subjects cook most of their meals (as opposed to eating out) and represented a diversity of income and age brackets.