Weeks 8-11 (June 20 - July 15)

Wow, I haven't updated this is quite a while! Sorry, I guess I let it get away from me. I am setting a weekly reminder right now!

So last time I posted I had just started to do the initial interviews for the diary study. Since then, I have finished conducting the initial interviews, and I will actually complete my final wrap up interviews today (wow, it really has been a while!). Most of my work lately (other than conducting the interviews) has been transcribing the interviews so we can run data analysis on them. We plan to create affinity diagrams to see what themes emerge, and go from there.

I also presented in our weekly Scrum meeting about the work that I am doing. It was a short presentation since it  was in the first few days of the actual data collection and I could only really present the plan and reason behind the study, but I think it went well.

Next week, I plan to continue transcribing. Hopefully I will finish that soon, so we can begin the analysis. The biggest roadblock there is simply that transcribing is tedious work, and I find it hard to concentrate on for any length of time. But it will get done, and hopefully sooner than later!

Week 7 (June 13-17)

Week 7 marked the beginning of the diary study. Starting on Monday, I began to conduct the initial interviews for the study - I completed 7 interviews this week with 8 more to go next week. The interviews themselves were intended to collect the user data (age, occupation, etc) as well as get the participant's baseline perceptions of the IIS's that we are interested in looking at. The participants are instructed to begin the diary study (ie filling in the web questionnaire) the day after the interview, so we are already getting some good data.

Between interviews, I have been transcribing the completed interviews. Since they are recorded, it is necessary to go back and type them all out, which can be very tedious. Transcription is definitely not my favorite task!

Next week, I will finish up the interviews and continue with transcribing. I will also begin moving the data from the database to a spreadsheet so we can analyze it more easily.

Week 6 (June 6-10)

In week 6, I did a pilot run of the questionnaire with a couple of people in the lab, just to make sure everything works the way it should, and see if there were any bugs or issues that should be fixed before the actual study. Only 1 bug was found, the use of single quotes or apostrophes in the text input areas through off the code, so I had to encode all strings before trying to insert the data into the database.

While the pilot was running, I began recruiting for the study. I printed out a poster advertising the study, and posted it up all over campus, except in the Computer Science building (we don't want any experts, just average computer users). Since it's a field study lasting 2 weeks, we are offering a $100 gift certificate for participation, and I think this factored quite a bit into the fact that I got quite a few responses in the 24 hours after the posters were put up.

I also began doing the write up of the experimental method for this study, which basically includes what kind of study we are doing, how we are running it, how we are recruiting. Pretty much, everything that I can write up without having any data. This study is going to be turned into a paper to be submitted to the International Conference on Intelligent User Interfaces, so having it written up early will help with making the paper as solid as possible.

Next week, I will be starting the initial interviews of subjects, and the study will begin!

Week 5 (May 30 - June 3)

I have completely finished the webpage (barring any unfound bugs), and I am super excited about that! I added a bit of color to the page, but it's still fairly plain. Anyone wanting to see it can go here, but PLEASE do not enter any information unless you are part of the study.

The pre-interview script is also completely finished, and the post-interview script is as finished as it can be without seeing any data. So that is what I did in Week 5. As you can see, I didn't have the busiest week, but the next few weeks to come will be full with interviews and transcriptions, so I enjoy the slow times while I can :)

Weeks 3 and 4 (May 16-27)

Firstly, I want to apologize for not updating this blog more often. I will try to keep on top of it! 

The last 2 weeks have been very productive! I did 2 iterations of pre-study interview questions, which are almost finalized. I have done 1 iteration of post-study interview questions, which need a bit more work. I was not as concerned with the post-interview because I had to get the webpage up and running, and to do that I needed to flesh out the questions that will be in the online diary, so I  did 2 iterations of those as well. 

As for the webpage, I abandoned Javascript and servlets in favor of HTML and PHP. I was having a bit of trouble connecting to the database the way it was, but after the switch everything is connecting great and the webpage is almost finished! It looks a bit plain, nothing fancy, so I might try to add a little cosmetic flair to it this week. This is the first time any web code I have written will be live, so its pretty exciting for me!

This week, I plan to do another iteration of the post-study interview questions, add some cosmetic improvements to the webpage, and make any changes to the webpage that are needed after Andrea takes a look at it. I might also look at how we will start recruiting for the study, since the interview questions and the webpage are almost finished.

Week 2 (May 9-13)

This week, I continued to explore the existing research on IIS's and explanations by reading 1-2 research papers on related topics every day. I worked on the diary questions and experimental design a bit more, but by the end of the week I felt like I might have gotten as far on that as I could without some feedback from Dr. Bunt, who is currently in Vancouver at CHI. I also started to work on the code for the website that we will be using in the diary study.

A lot of my focus this week went towards working on the webpage. Even though we don't have the details completely fleshed out yet I wanted to start this a bit early since I only have a little bit of experience in web programming, and I wanted to get a head start on (re)learning html, servlets, and Javascript in case we need to to anything dynamic. Specifically, I worked on things like login screens, and hooking up the webpage to a database. I did run into a bit of a roadblock here because I had an issue with accessing MS SQL Server from my computer, and I had to wait for the IT guy to have time to come help me with that. 

Week 1 (May 2-6)

I started at the HCI lab at the University of Manitoba on Monday, May 2. The first day was all about getting keys, accounts, and computers set up, and getting introduced to the project that I will be working on for the summer (see below for a description of the project I will be working on). The rest of the week, I was reading papers related to the project, and trying to get up to speed on the topics involved. I also had a decision to make regarding the avenue of research I wanted to take regarding this project - whether I wanted to focus on creating a diary study to gather more user data in a longer term field study, or focus on creating prototypes for how to give users the explanations of how the applications they are using make their recommendations. Ultimately, I chose to go the route of the diary study.

To get started, I needed to gather a bit more research of field studies completed on similar topics (of which there are relatively few). I also started to brainstorm some ideas of questions that we would need to ask users in the study, and started outlining the general experimental design.

My Project:

I will be working on extending some previous work that Dr. Bunt has worked on regarding Intelligent Interactive Systems (IIS's), specifically ones that assist users in making low-cost decisions, and how these systems are perceived and understood by their users. Dr. Bunt is also interested in discovering whether or not explanations of how these IIS's work are desired by users, and whether or not these explanations help users make decisions  based on the recommendations of the IIS, or improve user interactions with the systems themselves.

To further this stream of research, I am working on implementing and running a diary study. This will involve generating and refining the questions to be included in the diary, determining how long the study will run for, coding the online diary system, figuring out who we should recruit for the study (and where we should recruit), how participants will be compensated for their participation, and possibly performing some data analysis on the results if there is time.