Tuck-Rensselaer Research Project on Data Exploration

Welcome to the Tuck-Rensselaer website for research on how experts and novices explore data. Your participation in this study will contribute to our understanding of how people explore data, and ultimately will improve our ability to teach this vital skill. We ask that you devote no more than 30 minutes to the task of exploring a dataset. Please follow the directions below.

Imagine that you are a consultant to the Marketing Director of a small software company. The Director has hired you to assist her in assessing the performance of the company’s website as a marketing tool. The Director has provided you with the accompanying dataset (Web Visits.xlsx or Web Visits.csv), which contains information provided by Google Analytics on daily visits to the company's web site over a period of one year. In addition to the day of the week and the calendar date, the fields in this dataset are defined as follows: TotalVisits: The daily count of visits to the web site. TotalPageViews: The daily count of all web pages viewed by visitors. AvgTimeOnSite: The average time (in seconds) spent by a visitor on the web site. (Note that a visit to a single page, followed by an exit from the web site, is counted as 0 seconds.) BounceRate: The percentage of single-page visits, i.e., visits in which the visitor left the site directly from the entrance (landing) page without looking at any other pages. Your task for the next 30 minutes is to examine these data using any software you wish, creating whatever summary statistics, plots, or tables you think would be helpful to you, and report what you have found that would be interesting to the company.

 

Remember to speak your thoughts as you think them ("think out loud") without self-censorship. We are interested only in studying the thought processes you use in this task, not in the results you come up with.

Detailed Instructions

1. Click on the link at right to download either the Excel file Web Visits.xlsx, the text file Web Visits.csv or the data file Web Visits.Rdata
2. Go to the website for Screencast-o-matic, which will record a video image of your computer screen as you explore this data and an audio file of your voice as you describe what you are doing and why you are doing it.
3. Login to Screencast-o-matic, using this ID and password:
   ID: sgp@dartmouth.edu
   Password: 27Bradford
4. Click on “Watch a Very Quick Video” for a short introduction to this software.
5. Choose “Start Recording” (choose Run Java applet if necessary)
6. From the toolbar:
   a. choose your microphone
   b. set microphone volume
   c. choose Fullscreen
7. Click “Start Recording”
8. When done, click “Done.” (Be careful not to accidentally click "Restart".)
9. Choose “Publish to Screencast-o-matic.”
10. Use as Title: “Data Exploration Test” followed by your name.
11. Choose “Upload to SOM.” (This option appears in the bottom right corner of the window.)
12. Optionally, save the recording to your hard disk as a backup.

Final Reminders

• You may use any software you would like to explore the data.
• Remember to talk steadily about what you are doing, why you are doing it, what you have found, and so on.
• Use your mouse pointer to help us focus on what you are looking at on your screen.